Document zoXpJ7zbowqw5MBepw4vaVJR0

INTEROFFICE MEMORANDUM Subject Date 18 October 1977 Angiosarcoma in Great Britain ToDISTRIBUTION From. J. T. BARR (Location, orowitutton, or Datuttmant) (Location, Organisation, or Department) 01 ST: G. Baise, Beveridge, Fairbanks & Diamond J. Hadley, Keller & Heckman J. Lawrence, Society of the Plastics Industry, Inc. W. M. Smith, APCI T. R. Torkelson, Dow R. N. Wheeler, Union Carbide Corp. I enclose a copy of a report paper on angiosarcoma in Great Britain, and an associated press release, which I received today from Dr. John Stafford of I.C.I. Although written very cautiously, it would seem to support the absence of risk to the general population from vinyl chloride. As a side light, an Incidence of a little over one case per year for 50 million persons is a lower rate than the estimated 25 per year per 200 million in the U.S. JTB:djm (320) AP00008404 BRmSH medioL journal 8 OCTOBER 1977 PAPERS AND ORIGINALS Angiosarcoma of the liver in Great Britain, 1963-73 PETER J BAXTER, PETER P ANTHONY, RODERICK N M MacSWEEN, PETER J SCHEUER Hruitk Mtdical Jour>Mift 1977, 2, 919-921 Summary Death* attributed to primary angiosarcoma of the liver (ASL) in Great Britain between 1963-73 were reviewed by submitting available histological material to a panel of hlstopathologiste and by obtaining full occupational and residential histories for the cases agreed as ASL by the panel* On average four recorded cases of ASL occurred a year, but in only one-third of the cases submitted did the panel agree with die original diagnosis. Only one of the agreed cates could be confidently associated with exposure to vinyl chloride. Introduction Studies of polyvinyl chloride (PVC) workers have been set up in several industrialised countries after the recognition in 1974 of the association between primary angiosarcoma of the liver (AS!.) and occupational exposure to vinyl chloride monomer (VCM), the chemical used to manufacture PVC In Great Britain the Employment Medical Advisory Service (EMAS) agreed to undertake a mortality study1 of workers engaged in PVC manufacture and a case-control study* of all known cases of ASL occurring from 1974 onwards. As a preliminary study, we reviewed deaths attributed to ASL in 1963-73, in which available histological material was examined by a pane) of pathologists, and full occupational and residential histories were obtained in those cases agreed as ASL by the panel. Material and methods In 1974 the Rcgistrtrs General of England and Wales and of Scotland weft requested to send to EMAS i copy of si) death certifi cates from 1963-73 (Scotland 1965-73 only for administrative reasons) that recorded any type of primary sarcoma or angioma of the liver as a cause of death. These certificates were sorted by one of us (PJB) into two groups--namely, ASL and other Ever sarcomas. Certificates were categorised at ASL if any of the following diagnoses were mentioned, irrespective of whether the cates were infants or adults: haemangio- or angiosarcoma, KupfFcr-ccll sarcoma, {Hemangioendo thelioma, haemangioblattoma, or malignant angioma. Other cases of ASL were sought by placing notices in medics! journals and in the news-sheet of the Royal College of Pathologists, and by undertaking a search of the published reports. Hospital pathologists were invited to forward to the panel of pathologists histological material from the ease* of ASL and other liver sarcomas, which were to be used as controls. The panel regarded the following histological features as being helpful jn making a diagnosis ofASL: (1) an essentially intrarinusoidal pattern of growth, though solid and cystic areas may also be present; (2) cytologica) findings compatible with Kupffer-cel] or endothelial cell origin; (3) characteristic retieulin pattern of dense intratinusoidal network; (4) diffuse involvement ofthe affected part of the liver. Other Health and Safety Executive, Employment Medical Advisory Service, London W2 4TF PETER j BAXTER, msc, Mtcr, employment medical adviser Middlesex Hospital, Loadeo WIN tAA PETER P ANTHONY, mb, mkcmth, senior lecturer in pethelofy Western Infirmary, Glasfaw Gil BNT RODERICK N M MacSWEEN, net. 14*CT*th, reajier in pathology Royal Free Hospital, London NWS 2QG PETER J SCHEUER, mo, ncmth, professor of dinkal hittopttholes? HOl---Anruiilincidenceofallrecordedctaetof angiwrcoio* of the liver (ASL) in England and Wales 1963-73 and Scot land 1965-73. AP00008405 20 let* important feature* wm octwncdullaty fciemopoissi*, Otythro- phagocytosis, a tendency to thrombosis, Infarction, and touring. The panel was milted to classify the histological materia) at follows. ASL--if the reader war confident about the diagnosis of ASL. Not ASL--if the reader was confident that the sections were from a control or from case wrongty labelled as ASL. Doubifut--if the rwader thought the section might be ASL, hut was not confident that other diagnoses could be excluded. Unclaisifiabk--caret In which the material was inadequate or un satisfactory, or casts in which the reader considered that additional data would not enable a definite histological diagnosis to be made. The same slides were circulated in turn initially to four, subsequent ly to three working members of the panel, who examined diem independently and without knowledge of the original diagnoaea. If aD the members of the panel speed on the diagnostic category then dlls teat accepted. When there was disagreement sections were projected at a joint meeting of panel members and discussed without knowledge of the previous diagnoses, after which tbe consensus view was taken. Occupations! histories, including placet of residence, were sought for all the agreed cases. These were obtained, when possible, fom an interview with the next of kin by an employment medical or adviser using a standard questionnaire, and from factory medical officers. BRITISH MBDICAL JOURNAL % OCTOBER 1977 differentiated or, in oea ease, because of insufficient material. 'Section* * were received for one of the ialtatt only end wt etanfced u not ASL. At ths panel classified so few cases aa ASL it seemed that the as recorded on tbe death certificates might not accurately rcficei tbe necropsy findings. To assess this we studied the necropsy detail* available for 25 of tbe SO eases notified by death certificate that the panel saw. AS tbe necropsy diagnose* had been based on histo logical examination (five on biopsy specimens only) sod were found to correspond to those on the death certificates with only two caceptfens. CONTPOU Twenty controls, 15 cases of sarcoma to which five of carcinoma were added, were also seen by the panel. AS of these were confirmed as not ASL with one exception, which was classified as "doubtful.** Tha cause of death for this control according to tbe death certificate was primary haemangioma of the liver, whereas the necropsy diagnosis was4`anaplastic primary cancer of the liver." Result* CASES For the period! 1965-7) in England and Vales and 1965-75 in Scotland 41 death certificates were received which mentioned a diagnosis of ASL. There were 27 males and 14 females, including one female and two nude infants. Two patients, both men, died la Scotland. The search revealed nine published cases*'* and a further case was reported to BMAS. AH these patients had died between 1967-72. The reported case and five of tbe published cam would otherwise have gone undetected beeause the diagnoses mentioned on the death certificates were other than ASL. The annual incidence of all the recorded cases rote to a peak of nine cases in 1968, hilling to three cases by 1973 (fig 1). Histological sections were obtained for 36 (77%) of the recorded cases. The panel agreed the diagnosis of ASL fat only 14, and classified seven as doubtful and twelve as not ASL (table I). Three other cues were regarded as unclasrifiabk either because the tumour was poorly TABLE I--Ctotsi/utitieH by pantl qf polhelegisn ecu* j ASL Mcum'v in EtgicnJ and Watti (7P63- 7J) end Scotland (IW5-73) No of petientt Nu even by panel ,, Ail.......................... (.itwlamrinhlc .. Not ASI.................. Death cere ficate men Male Fmnelc ST 14 20 10 11 3 1 97 Other cases* Male Female S3 1 'Published ciki i ihou not I'btarned from death certificate*. ASI, pruuarv angiosarcoma of the liver. 36 13 cam aoubo by ns panel The death certificate dhgnosrs and other detail* of the 14 cates of ASL agreed by the panel are given in tabic 11. There were only two women. In throe published cases the underlying causes of dead) were recorded as ovarian cyst, hepatic failure, and ischaemic heart disease respectively, indicating that cases of ASL will be missed if reliance is placed solely upon data from death certificates. The recorded disposes for the other cates show that confusion may arise over the nomenclature of ASL, as rix dlffbrait terms for ASL arc shown. Occupational histories were obtained for 12 of these eases. One had been a process worker in PVC manufacture exposed to levels of VCM exceeding 200 parts per tuition (ppm) for over 20 years.3 * The glass- fibre worker who died aged 56 in 1970 had been employed for 11 years until the age of43 in a factory which manufactured rubber components and coated fabric with PVC. The latter process took place in a small pan of the factory, and although there was no evidence that he bad actually worked there, the possibility could not be excluded. There was no indication that any of the other cases had worked in the manufac ture or fabrication of PVC or with other chemicals. Commercial production ofPVC began in Britain in 1941. Four main tftei were in commercial production before 1954, and two others in 1969 and 1971. The locations of these factories and their relation to tbe last place of residence of the 14 cases are shown in fig 2. Eleven cases had either been bom or had worked for many years in tbe area In whieh they were recorded as dying, but this information was ooi known for tbe other three. The cases were widely scattered except for two fat the north-west of England. One of these was the process worker described above j within half a mile of the factory where this man had worked, the other patient had lived for seven years before his death in 1970, having spent 40 years as a bus driver before becoming s cashier. One case, who h*<4 been a hospital attendant, had undergone a carotid angiogram in 1959 in which thorium dioxide 'Thoronast, had bean used.* Other medical details--for example, drug usage--were not table i!--Detail! of all ceset ciauifM by tHt pcml m mgictcrctmc tf the liver OiatnoiU according ie> death certificate Angiosarcoma .. Ourlin cyst .. Haiirafljpoi'lKjulhclioBii Malignant hecTnengiBna Hecfnanginblxiworas.. HacmaociocnJiithelioau .. Haansngiocndvlheliotnj Hepatic failure HectnsnpocnUolheiiomi .. leehaemle Seen disease HicmaneiocnJotheliam* .. HicmuntioendmhtliomennM Angiosarcoma KuplTcr-cell sarcoma VCM vinyl chloride monomer. PVC polyvinyl chloride. Year of death 1966 196? 1968 1S6S 1062 1669 1970 1970 197] 1971 >972 1972 1973 1973 Sax M r M M M M M M M F M M M M Age a death (yean M so 64 SO 70 1 36 70 *4 71 69 46 49 Occupation according to doth certificate Sleetnoiet engineer Hetucwifa Electrical engineer Bricklayer Miehinery Alter Railway porter Cathler Fibre (Ue> worker Rubber planter Kouamfe Process worker Hospital attendant Tailor Lorry driver Exposure to VCM Not known No Not know n No No No Lived neat PVC plant Possibly PVC paste No No Ye* ^boritun dioxide) No AP00008406 tYwT - f-' Ac:?, in ilC rifN l Njr f'ai i>or7 I- Vc iTlu v*v- ' ) { V.':\n It.I? r>|i t. liov.-w News Release Health & Safety Executive Vtyntrdft Hama I Chepstow Piece TtUpftont 01-229 3436 to. 179 October 7* 1977 ary nr uvga cxfcer FROM VCtt oonm have bhm ovrasmiiap bats mow A preliminary report publiahed today suggests that the hshard fro* the chemical YCK to workers ewployed in. the manufacture of tVO or PVC products and to the general public Hiring arcusd the plants involved not ho as great as v&s first exposed. VCM (vinyl chloride eenoser) is the gee used la the aaaufnature of polyvinyl chloride (PTC). She report, published today (October 7) in the British Medical Journal, is written by Dr. Peter Barter of the Staltfc and Safety fcieetttive** Bq^loywant Nodical Advisory Service (DIAS) sad three other aedleal experts. In 197nediesl findings fron the 08 indicated for the first tias that workers engaged in the Manufacture of PTC were at riak fro* developing liver cancer .(angiooarcoos) Urgent action by the Health and Safety JhMMhztlve in Collaboration with both aides of the eheaical iniustry resulted In levels Of exposure to TON being drastically reduced in the six British plants . involved. the report says that oat of * total of V? reported eases of aagiosttaoMa of the liver in the ten-yesr period 1963-73, in only Ik was the verified end out of those only one could be confidently associated-with join^nsnre to VCM. It also says that the overall incidence of BHglbiartMa did not appear to be geographically listed Ht the lsaattsa of factories dealing with the tffasnlcal' although there wao one oonftraed ease Of shgiottfeowa la a aaa who had lived within half-a-aila of a PTC factory for Sit years before his death. There waa no indication of snglOBarcoaa having Arisen through swallowing aineta saounts of VCM, for exaaple, froa plastic feed wrappings. AP00008407 -2- ... Out of k7 deaths la the period which were attributed to angiosarcona* n*y the report* 37 Had sufficient information supplied by hospitals for the date to be scrutinised by the peael of medical experts. Where the ceuae of death would be definitively screed as angiosarcoma (2% Poses), full occupational and residential ease histories were obtained and analysed in 12 rinse hert-gf-kin could be traced. Ih the one oase risere workplace exposure could be oeeceieted with the worker*# fatal illness* he had been exposed to VCK levels probably exceeding 200 ports per Billion ih air for over 20 years. (The current agreed level in this country for worker exposure which is enforced by the Factory Inspectorate ia ten ports per Pillion in sir ever on eight-hour day.) The authors of the paper point out that their findings ore generally reassuring in that not only does the annual incidence of angiosarcoma appear to be in single figures* in a population of about 50 Billion* hut exposure to VCK sees* to have been responsible for only one death over tbs whale ten-year period. However they say that there renalna a possibility that angiosarcosa nay be associated with VCM at low exposures so a close watch en the srnnrtetwm of this tumour should be naintained. The paper* "Angioesreosa of the liver in Great Britain* 1963-73"* gives the results of s preliminary study* which is now being extended to include ell known eases of sngiosarcocs occurring iron 197** onwards. This survey is being undertaken by DUS* ae is a full survey of all causes of death among workers engaged in FVC manufacture, this.work parallels studies set up in a number of industrialised conntries sines 1974. Ad DUS spokesmen eoBaa&tedi "These findings give ns pounds to hope that this ehssdonl will turn out to be less of a cancer risk than wan originally supposed when the American findings were announced. The single sake of t aan living near a WC factory who died froa eagteaar--a ia insufficient evidence to draw any conclusions at this stage." j # *S ! AP00008408 ncfffcs ip ttrroKS: 1. the text of the report is attached. JL 5be period 19S5-73 was chosen for study in order to discover the frequency of angiosarcoisa of the liver* and to evaluate any trends on it* incidence before the tine when e relationship between occupational exposure to YUM end liver cancer became known in 1974. 3. A first report of the full surrey of all causes of death In sen engaged in PTC manufacture In Oreat Britain was published in the British Journal ef Industrial Medicine in February 1977* Zt shoved that up to the and. of 19?V there were twe deaths from angiosarcoma of the liver involving workers exposed to VCN: ene s aan who died in 19711 is included in the present report; the other occurred in 197b sad is outside the study period of the present report. Bines than no other eases of angiosarcoma, are known to have occurred stone British worker* engaged in PVC sanuXaoture. f \ ' B69:77 AP00008409 BRITISH MEDICAL JOURNAL 6 OCTOBER 197? Fie 3--Mein sices of PVC manufseturt in Grear Britain and tas place of retidtnct of the casat clitsified at ASL by the paneL sought for the 1963-73 cases, but EMAS intends to obtain these for all coses occurring from 1974 onwards. Discussion ASL is a rare tumour, and v/c expected that there might be a (airly wide discrepancy between the original diagnoses and those reached by the panel. Our findings show that between 1963-73 an average of four cases were recorded annually, and In only one-third of the cases submitted did the panel agree with the diagnosis. Undoubtedly the histological material seen by die panel may sometimes have been inadequate compared with that available when the diagnosis was first made by the hospital pathologist. Nevertheless, in one-third of the cases the panel was confident that the diagnosis was not ASL, showing that this explanation was inadequate for these cases at least, We know of only nine published cases in which ASL was reported as the cause of death in Britain before 1963 {Scotland 19o5). ` '* One of these was attributed to gamma radiation received from a radium needle that had acddently become lodged in the myocardium of a patient,* and another case was of suspected idiopathic haemochromatosi*,5* Thorium dioxide (Thoroirast), a radioactive contrast material introduced in 1928 for angiographic nudies, has been the agent most often associated with ASL, but we came across only one such case in this study. By 1950 thorium dioxide was used infrequently in Britain} it had a piece in the radiography of cerebral abscesses until the late 1960s, usually remaining localised to the site of injection, and rarely entering the circulation.11 In other countries ASL has been associated with arsenic exposure--in vineyard workers contaminated from the use of arsenical sprays," and in patients with chronic arsenic intoxication after prolonged taking of arsenical drugs.'* '* ASL has also been recorded in a vineyard worker who had worked with copper-containing sprays (Bor deaux mixture) for many years,'* In Britain the use of arsenic in pesticides and agriculture generally has diminished consider ably since 1960, and the use of arsenical drugs was largely dis continued after 1950. 921 Since 1950, however, the PVC industry* has developed rapidly, the material having become one of the most widely used plaatica. About 8000 people have been employed it some time in PVC manufacture, over halfof whom have been exposed to VCM levels exceeding 25 ppm in the factory atmosphere.1 Present mean levels of VCM in the atmosphere up to half a mile way from the factory where the neighbourhood case of ASL occurred are considered to be between 0*01 to 0*03 ppm, and may have been several tunes higher in the past, indicating the order of exposure of people living dose to factories manu facturing PVC (In 1975 a code of practice was introduced, and this was accompanied by a major reduction of VCM levels in and around factories manufacturing PVC.) About 35 000 people in England and Walts art currently working in factories making PVC products, where exposure* have been much lest than the current hygiene standard of 10 ppm (HM Factory Inspectorate's Records). In addition the general population has ingested ex tremely small amounts of VCM which have entered foodstuffs from PVC wrappings.1* Despite these observations, only one case of ASL occurring in 1963-73 could be confidently attributed to VCM. One patient was suspected, hut not proved, to have worked on a process which used PVC as a raw material, and another had lived for tlx yean within half a mile of a plant manufacturing PVC. These case* hid been diagnosed a* ASL by hospital pathologists several yeert before the association between VCM exposure and ASL in man tad become known. Although our findings are generally reassuring, in that the annual incidence of ASL in a population of about 50 million seems to be in single figures, the possibility remains that ASL may be associated with VCM at low exposures. Thus more people may be at risk than is currently supposed. Our data provide no reason to suppose that most of even those fow cues that have occurred are pro duced in this way, but dose watch on the occurrence of this tumour should be maintained. We gratefully acknowledge help from the Registrars General for England and Wales and Scotland; the Medicines Division, Depart ment of Health and Serial Security; HM Factory Inspectorate, and HM Alkali and Clean Air Inspectorate. We thank the ftetory doctors and the employment medical and nursing advisers, who obtained the occupational histories; Prof H K Wembren, who was a member of the panel of pathologists at the beginning; Dr* $ Gauvaio and M Greenberg for their advice; and Mr G Sparkei for administrative support. We are especially grateful to hospital pathologists for sending histological material. Reference* I Fox, A J, and Collier, F F, British Journal of Industrial Mtdieins, 1977, 54,1. * Baxter, P ], Procttdsnfs sf At Beyat Soatiy of Mtdiant, 1976, 69, 297. * Barks, J, Hamer, D B, and BJumglTt, L H, British Journal of Clinitsl Prti, 1972,54, 261. 4 MacSwecn, R N K, n dl, Journal of Pathology end Bacteriology, 1973, tee, 39. * Campbell, J W. and Webb, J H,Journal tf sAt Boyal Ceikgi of Surgtcns 4 fid'efeir/k, 1914,1*, 233. * Lee, T I, and Harry, D S, Lasutt, 1974, 1,1316. * PoOard, S M, and MiBward-Sedler, G H, Journal tf CKritol Pathology, 1974, 53, 214. * Gay, },Journal tf Pathology and Bacteriology, 1929, 35, 337. *Rou> J id.Journal tf PasMegy and Bacwiolegy, 1933, U, B99. II OgUvit, R r, tad Mackenzie, I, Journo! 4 Pathology and Bacteriology, 1936, 43,143. i Hailings-Jimet, R, Journal PathMegy end Bacteriology, 1949, 61, 49. 11 Biker, H de C, Pafet, G B, and Davson, ], Journal tf Pathology and Baeurietco, 1936,75,173. ** du Boulsy, G, personal comnuiniattioa. 14 lUfelson, W, at al, Cantor, 1965, 31, SJ4. u Lander, J J, i al, Ganroanltrolagy, 1973, 61, 1532. 14 Pimentel, ] C, end Mtaun, A P, Oostrotnttroiegy, 1977,12, 27S. 'Bernes, A W, Proceedings ef tht Royal Seritsy cf Mediant, 1976, 49, 277. (AttefSod If August Jt?T) AP00008410