Document G8omgQdZ56dEmgd11M0w1ndY

FILE NAME: Welding (WELD) DATE: 1971 DOC#: WELD010 DOCUMENT DESCRIPTION: Journal Article - Diffuse Malignant Pleural Mesothelioma and Asbestos Exposure T H THE O R A EDITED FOR THORACIC SOCIETY ( P r e s id e n t, i, c;. s c a d d i n g ) X by R. ABBEY SMITH COLIN OGILVIE ASSISTANT EDITORS (Medica!) C. P. Aber D. G. Julian (Surgical) D. B. Clarke EDITORIAL COMMITTEE T. Anderson C. M. Fletcher R. Marshall X. R. Barrett J. C. Gilson J. G. Scadding J. Leigh Collis D. A. Heath H. Spencer F. Ronald Edwards A. Logan O. S. Tubbs A. Young (Statistical Adviser) Editor, British Medical Journal VOLUME 26, 197! BRITISH MEDICAL ASSOCIATION, TAVISTOCK SQUARE, LONDON, WC1H 9JR 1 ' i Thorax (1971), 26, 6. Diffuse malignant pleural mesothelioma and asbestos exposure F . W H 1TW ELL and R A C H EL M . RAW CL.IFFE Rroadgrcen Hospital, Liverpool ; Aintrce Hospital, Liverpool ; and St. Catherine's Hospital, Birl etdicad Pleural mesothelioma has been diagnosed in 52 patients in three hospitals on Merseyside between 1955 and 1970, 60"u being diagnosed from operation specimens and the rest from postmortem tissues. Necropsies eventually held on nearly half the operation cases conlirmed the diagnosis, giving a necropsy rate of 70% for the series. The morbid anatomy conformed to earlier descrip tions except that widespread mtastass were much commoner than has usually been described. Histological findings agreed with previous accounts of the tumour, except that, in our hands, special acid mucopolysaccharide staining was less reliable than Southgate's mucicarmine, which was of value in differential diagnosis. Association with asbestos was confirmed from industrial histories in 80 v, of cases, the commonest industries involved being shipbuilding and repairing in men and sackware repairing in women. Lungs of industrial mesothelioma cases showed basal asbestosis in 17%, and excessive asbestos bodies in almost all the rest. Quantitative comparison of asbestos bodies in lung smears from mesothelioma cases compared with lung smears from other Merseyside adults showed much higher counts in the mesothelioma eases. The interval from first exposure to asbestos until appearance of mesothelioma ranged between 13 and 63 years, with a mean of 42 years. We think the incidence of mesothelioma will continue to rise with the increased use of asbestos until about 40 years after adequate protective measures have been taken. The recent reinstatement of diffuse malignant pleura! mesothelioma as a legitimate neoplasm dates mainly from the papers of Godwin in 1957 and McCaughey in 1958. However, the con troversy as to whether these pleural tumours are primary or almost always secondary might have remained a nice academic exercise but for the work of Wagner, Sleggs, and Marchand who in I960 reported an association of pleural meso thelioma with asbestos exposure. They described 33 pleural mesotheliomas associated with asbestos exposure in Cape Province, the exposure being more frequently environmental than occupa tional. Since then there have been many studies link ing both pleural and peritoneal mesotheliomas with asbestos. In the United Kingdom the largest series of cases published have occurred in Hast London, usually in people working in or living near asbestos factories (Emieknap and Smither, 1964; Hourihane, 1964; Nevvhouse and Thomp son, 1965), and other accounts have described cases occurring in the major shipbuilding areas. The Belfast cases described by McCaughey in 1958 were reviewed in 1965 with further eases added, showing asbestos exposure in 32 out of 42 cases (Etmes, McCaughey, and Wade, 1965), In 1964 Owen reported 17 cases of mesothelioma on Merseyside with an asbestos link in 82 % of cases, and in 1968 Harries analysed the trades at risk in the Royal Naval dockyards. These investigations have been largely retro spective, often involving much environmental and occupational research made after the patients' deaths, and also hampered by an inadequacy of preserved specimens, particularly lung tissue, making assessment of the asbestos association very difficult, which may as a result have been under estimated. The present paper is a prospective continuation of the Merseyside series reported by Owen in 1964, when he described cases occurring between 1955 and 1963 in what was also largely a retrospeeiive study. n i t SUBJECTS STUDIED Fifty-two cases have been examined, comprising 14 of the 17 cases described by Owen 0 964) and 38 addi tional cases which have occurred between 1963 and 1970. The three cases of the earlier series not included are one which was a peritoneal tumour, one for which there were no histological sections, and one this case 6 14) for which the Alt but one pa have been patient Hospital, Liverpi Birkenhead. The of Merseyside pU ate referred to th at Broadgreen Catherine's Hosy of those who wo all the cases wen died in other ho times performed were available It times no lung tr, HISTOLOGICAL i The following r< hospitals for thi ourselves in 18 t * an: tumour N \ and its direct ext \ dunes, were cxw i Sections were star ` coon. Southgate : ihe earlier cases i acid (Wagner, 5' i colloidal iron, U ' and periodic acid >treated and unli ' abandoned as tl nine to formol-sa L % i;| l.UNGS The honeycomb arc; 'examined from bioes of a low 'normal. In addit; the base of a lo~ bodies. t.l NCI JUICE SN readily demonst juke than by e L method oE lung * out this study m ! adult necropsies , n t c method is I been described r ensitivity as the Glasgow, and / \ provide a comp> , tent of the urba of the mesotheh At necropsy removed and e t squeezed hard f collects. Enough the red eel Diffuse malignant pleural mesothelioma and asbestos exposure 7 14) for which the records and slides have been lost. All but one patient, who was in the earlier series, have been patients in Broadgreen Hospital or Aintree Hospital, Liverpool, or in St. Catherine's Hospital, Birkenhead. These three hospitals admit the majority of Merseyside pleural mesotheliomas, because patients are referred to the Regional Thoracic Surgical Centres at Broadgreen and Aintree Hospitals, while St. Catherine's Hospital is the district hospital for most of those who work in Birkenhead shipyards. Though all the cases were diagnosed in these hospitals a few died in other hospitals where necropsies were some times performed. Tumour tissue and necropsy reports were available for study in these cases, though some times no lung tissue was examined microscopically. pulmonary oedema is present. The beaker is left at room temperature for a day, when the contents are transferred to a conical centrifuge tube. After spin ning the tube the supernatant is discarded and smears of the deposit are spread on a microscope slide. The dried smear is mounted in balsam and examined with a 2 3 in objective. Quick identification of asbestos bodies can be aided by staining with Peris' stain, but the intense blue staining sometimes makes it difficult to differentiate true from pseudo asbestos bodies, as the centre of the body becomes obscured, so tinstained smears are preferred. 'This method of preparing smears can be used quite satisfactorily with lungs which have been fixed for several years. HISTOLOGICAL EXAMINATION IN NECROPSY CASES DIAGNOSIS The following routine has been adopted in the three hospitals for the last 24 cases, being performed by ourselves in 18 cases. the tumour Several tissue blocks from the primary and its direct extensions, and blocks from any secon daries, were examined microscopically in each case. Sections were stained routinely with haematoxylin and eosin, Southgate's rnucicarmine, and alcian blue. In the earlier cases, in order to demonstrate hyaluronic acid (Wagner, Monday, and Ilarington, 1962), Hale's colloidal iron, alcian blue, Southgate's rnucicarmine, and periodic acid Schilf were used on hyaluronidasetreated and untreated sections, but the method was abandoned as the results were inconclusive, possibly due to formol-saline fixation. the lungs The lungs were examined for line fibrosis, honeycomb areas, and scarred areas. Blocks were examined from any such areas but also from the bases of a lower lobe when ihe lung tissue looked normal. In addition 20 /i thick unstained sections from the base of a lower lobe were examined for asbestos bodies. LUNG JUICE smears Asbestos bodies are more readily demonstrated by examining smears of lung juice than by examining lung sections. The same method of lung juice preparation was used through out this study and also in a series of 200 consecutive adult necropsies at Broadgreen Hospital in 1966-7. The method is not as searching as many that have been described but seems to have about the same sensitivity as the techniques used by Roberts (1967) in Glasgow, and Ashcroft (1968) on Tyneside. It does provide a comparison of the lung asbestos body con tent of the urban population of Merseyside with that of the mesothelioma patients. At necropsy a basal segment of lower lobe is removed and cut into several pieces, which are squeezed hard over a beaker, into which the fluid collects. Enough water is added to the juice to haemolyse the red cells, though this is unnecessary when The diagnosis has always been made from histo logical findings, obtained dining the patients' lives in 60% of cases and at necropsy examination in the rest. Only rarely was it possible to make a definite diagnosis from a needle biopsy specimen, and most tissues were obtained at thoracotomy when pleural biopsy, pleurectomy, decortication, and occasionally pleuro-pneumonectomy was per formed. The same staining methods were used with operation specimens as with the postmortem tissues. Positive diagnosis was made in 30 patients, confirmed eventually at necropsy in 14 cases. Eleven patients have died without necropsy studies, and live are still alive who have been diagnosed in the last three years. Tt was possible to find asbestos bodies in only four of these surgical specimens, which usually showed just narrow rims of subpleural alveoli, or consisted only of parietal pleura. In 22 cases the diagnosis was first confirmed at necropsy, though often it had been suspected clinically for months or even years. In seven of these cases there had been negative pleural needle biopsies. histological features The wide range of microscopical features re ported by many writers and well described and illustrated by Godwin (1957), McCaughey (1958), and Hourihane (1964) were present in this series. Though some admixture of types was found in most cases, the tumours were predominantly of one form in over half the cases. They were classi fied as follows: Tubulo-papillary Sarcomatous Undifferentiated polygonal Mixed 20 cases 11 cases 3 eases 18 cases 8 F. Wlntwell and Rachel M. RawcUfje Tumour giant-cells were sometimes present in the sarcomatous, polygonal, and mixed types. tubui.o-papii i ary rvPE These tumours can often only with ditliculty be differentiated from secondary adenocarcinomas, particularly those originating from primaries in the lung, prostate, stomach, breast, and ovary. The following features are seen in most meso theliomas of this type arid are much less com monly found in secondary carcinomas: 1. Tumour cells are regular in size and arranged in an orderly pattern. The nuclei are vesicular and regular, seldom showing mitotic activity, though there are exceptions to this last point. 2. The branching acini are lined by low columnar or cuboidal cells, which in some places tire so flattened that the acini look more like empty hyperplastic blood capillaries than glandu lar acini (Fig. 1). 3. The acinar cells and lumina rarely contain any secretion which stains brightly with South gate's mucicarmine, like epithelial mucin, though there is often secretion in the acini and within lining cells which stains vigorously with alcian blue. 4. Dilated acini often contain and may be dis tended by branching papillary fronds of tumour cells, which arise from acinar lining cells am cover, a tine connective tissue core (Fig. 2). Thi; feature is also seen in secondary ovarian carci noma but otherwise suggests a mesothelioma. 5. Where tumour has invaded Jung or lymph node the mesotheli.il tumour cells occasionali; are seen to contain dust or carbon particles in their cytoplasm (Fig. 3), which is very rarely .seen with carcinoma cells. : sarcomatous type (Figs 4 and 5) A wide range of appearances from cellular fasciculated fibro sarcoma through myxoma to large areas of rela tively acellular collagen may be found, but nuclei are usually regular and mitoses infrequent. There may be so much collagen and so little cellular tumour present that it is easy to mistake the pleural thickening for the plaques associated with asbestosis. In two eases only collagen was found in the tumour tissue encasing the lung, but where the tumour had extended through the pericardium into the heart and through the diaphragm into the liver, these organs showed typical fibrosar coma. FIG. 1. Capillary pattern in tubulo-pupillary tumour, liaematoxyiin atul eosin x 165. e disumour :> and This carci>a. i\mph lonally ics in v seen range fibroc rela nuclei There .ellular Te the d with found ; where . irdium <n into Vosar- ( t Diffuse malignant pleural mesothelioma and asbestos exposure ; *"\ T - ?- *Vv ,, ' >aV . f > * ,,' - , V/ . 'f/ -; . -.K/ Krr. - . T V ' ,J C ' v - i;Aa . . ^ *r ****!.* -- 1* * ^ , G 2 papuiary fronds in acini of tuludo-papillary tumour. Haematoxylin and eosin x 165. f F1,,. 3. Dust particles in cytoplasm o f mixed tumour, ,,aemmoxylin urn/ e,,M,, * 10 V) F. Whitwell and Rachel M . Rawclifje s ' ff ** is?:;-'". 'W ^ ' 4 y > ` Vi . * & * * . M i a>sin x 785. ** , >. - r . % la**.. * f 1 * / , , < ' ' +*r > /J / W m I ' 9 u < ' 1 + %* \ i) /* i ,1 * ' /> ' * 9t* , , : y/ . * " 3' I ' . ** "; t * ,, -^ 'f \ * / S ' / ^ >I f <* 1 * ' * - I k $ .v ` f A f>r 2 0 ^ ' *0*`* ^ <r - # -w w * 0 * 4*** *., at r *v O > t *> J-J * ,r > T T * ' * V , * > w . * . . . . . . * .. f i g . 5. Collagenous fibrosarcoma. Haematoxylin and eosin x 3X5. The tun sarcoma librosareon niycMiircon.. tibrosarcom I NDIFFERI this type m . it occurs ii described thelial. In r in three c tumour We: shows a ii have able When the their cell honoycorm where the cell outlin. cells form common. These si in most i differentia: distinguish of the plci infarcts), i biopsy nia` Diffuse malignant pleura! mesothelioma and asbestos exposure 11 The tumour must be difiercntiated from fibro sarcoma of the lung and from secondary fibrosarcoma. However, apart front uterine libromyosarcomas, the primary sources of secondary fibrosarcomas are usually clinically obvious. Very occasionally there occurs a much more malignant-looking variant of this type, with nuclear and cytoplasmic irregularities, mitoses, and tumour giant-cells, which it is ditlicult to dis tinguish from carcinoma. UNDIFFERENTIATED POLYGONAL TYPE Tissue of Ihis type is present in many tumours, but at times it occurs in an unmixed form, when it has been described as sheet-like, polygonal, or solid epi thelial. In this series it was found in a pure form in three cases, but large areas of this type of tumour were present in six other cases. The tissue shows a uniform pattern of regular cells which have abundant eosinophilic cytoplasm (Fig. 6). When the tumour cells are compressed together their cell membranes produce a polygonal or honeycomb outline like pavement epithelium, but where the cells are separated by tissue fluid the cell outlines become rounded. Occasionally these cells form tumour giant-cells, but mitoses are un common. These sheets of cells look remarkably benign in most cases, and the problem is less one of differentiating the lesion from carcinoma than of distinguishing it from benign reactive hyperplasia of the pleura (such as occurs over old pulmonary infarcts), though this dillieulty arises only with biopsy material. m ixed t y p e (Figs 7 to 13, all taken from one case to illustrate the wide range of features.) The mixed tumour is one of the commonest, and is the easiest to diagnose when adequate tissue is avail able. With this type of tumour there may be large areas where only one histological pattern is seen, so many sections must be examined, but in a for tunately chosen area not only is there a mixture of tubulo-papillary, sarcomatous, and polygonal elements but the tumour cells show all stages of transition between these types. The features described as characteristic of the previous types are equally well seen in the mixed types. In mixed tumours it is quite common to find that only one tumour element is present in some of the metastases. Usually this seems a matter of chance but the tubulo-papillary elements occur by themselves in affected lymph nodes, and only the sarcomatous elements have been found in bony secondaries. Many tubulo-papillary tumours are associated with a dense deposition of relatively acellular collagen, either a stromal response to the tumour fig. 6. Undifferentiated polygonal cell type, in solid sheets on right and as separate rounded celts on the left. Haemutoxylin and eosin x 900. no. 7. M ixed type. Fasciculated fibrosarcoma area with scanty tabular inclusions. Haematoxylin and eosin x lfi5. or to asbestos, with a mixed i matous tumoui the pleura the glandular acini he ve ry like th A p rim ary ce duce a rather has a bronchi; elem ent o f the tum our of aim nant synovium confusion. The appearand pleural mesoth and show little lung on the ah lapse and so secondary tun rubbery, thick. bo;h visceral a> dose cavities > s' i f k imB &- / to ' I \SC b r ir Mi FIG. 8. M ixed type. An even mixture o f tubuio-pupillary and sarcomatous types. Haetna- nu toxylin and eosin x 165. Diffuse malignant pleural mesothelioma and asbestos esposare 13 or to asbestos, and this should not be confused with a mixed tumour. Likewise, where a sarco matous tumour invades peripheral lung through the pleura the encased alveoli become isolated glandular acini lined by cuboidal cells, which can be very like the appearance of a mixed tumour. A primary carcinosarcoma of the lung can pro duce a rather similar appearance but frequently has a bronchial origin, and the carcinomatous element of the tumour is usually squamous. A tumour of almost indention! features is a malig nant synovioma, but this L not likely to cause confusion. MORBID ANATOMY The appearances of the cheM contents in cases of pleural mesothelioma have been well described and show little variation from case to case. The lung on the affected side shows compression col lapse and sometimes bronchopneumonia and secondary tumours, and it is encased in a rubbery, thick, yellowish-grey tumour involving both visceral and parietal pleura, which may en close cavities filled with necrotic or fibrinous material or glairy fluid. The tumour usually en cases the whole lung, but when not so advanced it may be possible to identify pleural plaques. However, these are usually present and best seen on the parietal pleura of the opposite lung. The tumour spreads into interlobar fissures and interlobular septa ; there is often direct spread to the pericardium and also, through the diaphragm and through the chest wall to subcutaneous tissues. Mediastinal and hilar lymph nodes are frequently affected, and there may be a diffuse posterior mediastinal spread with tumour encasing the aorta and oesophagus. The following features were found in 32 necropsies. pleura In five cases the pleural involvement was bilateral and of similar extent on both sides. In 14 cases the left side was predominantly or solely affected, and mesothelioma was found on the right side in 13 cases. p er ic a r d iu m and iimart The pericardium was involved by direct extension in 15 cases. T his was found in 10 of the 14 cases in which only the left no. 9. M ixed type. Mainlv sarcomatous but with two acini on the right side, anti a mitosis in the sarcomatous tissue in the lower centre. i/aemati>sylin and cosin x 3X5. l ia. 10. M ixed type. Myxomatous area with tumour cells attempting to form acini in the upper half, and looking sarcomatous in the lower half. Haematoxvlin and eosin :< 3S5. f K,. 1!. M ixed type. Capillary appearance in acinar area. Haematoyxlin and eosin f i g . 13. M ixed type. Tumour deposit in hilar lymph node showing only Itihnlo-papil/ary pattern. Haematoxylin and eos'm x 785. F. Wltihcell and Rachel M. Rawclil)e pleura was affected, in only 3 of 13 cases in which only the right pleura was affected, and twice when pleural involvement was bilateral. The heart was the site of tumour in 10 cases, always when the pericardium was also affected and usually by direct extension to the atria, but in three patients mtastass were present in the ventricular myocardium. All histological types spread to the heart with equal facility. i.y.viph nodes In 14 cases hilar and mediastinal lymph nodes were replaced by tumour. Axillary, cervical, and abdominal nodes were each involved on three occasions, and an inguinal lymph node once. Heard (1966) has reported lymph node in volvement to be uncommon, but in the present series it was found in 50%, of cases. It was noticeable that sarcomatous elements did not spread into lymph nodes though they often encased them, and the same applied to the sarco matous elements of mixed tumours. This agrees with the findings of Hourihane (1964). CHEST WAIT, OPERATION SITE, AND DIAPHRAGM Direct extension of the tumour into the chest wall on the same side was present in all cases, with tumour eroding the ribs. In 11 cases there were subcutaneous tumour nodules on the outside of the chest wall lixed to the deeper tissues. In six cases there was tumour spread along the track of healed operation wounds and needle biopsy incisions, but the incidence of this was probably higher as it becomes indistinguishable from the general subcutaneous spread. The diaphragm was invaded in nearly all cases, usually with tumour tippearing on the inferior surface, sometimes extending directly into the liver or right adrenal, but not often associated with general peritoneal extension. p er ito n eu m In two cases there was generalized involvement of the peritoneum and mesentery, and localized areas of mesothelioma were present in the peritoneum or mesentery in nine other cases. d is t a n t m ta s ta s s In the first 10 necropsy cases in the series, which were those reported by Owen (1964), there was only one case with a bloud-borne metastasis, which was in the liver. By contrast, in the next 22 necropsies, mtastass were found in 14 cases. The commonest sites for blood-borne mta stass were the liver and the lung, usually the opposite lung to that encased by the primary growth, though in some cases there were second- ' a ries in both lungs. There were also secondaries in the adrenals (3 cases), thyroid (2 cases), and kidney (1 ease). Bony secondaries were present in two eases, one in a lumbar vertebra and one ini right femoral neck. In one case there was i secondary tumour of the meninges in the right middle cranial fossa. All histological types of tumour gave rise to blood-borne metastases, but while over half the sarcomatous tumours did so, less than a quarter of the other types showed this spread. This changed incidence of metastases in the second part of the series reflects an increased confidence in the diagnosis and nature of the tumour, and the more detailed study of the pro spective cases in the series. Probably the retro spective series was incomplete through non- acceptance of metastasizing growths. Although Godsvin (1957) accepted extensive metastases in several of his cases without com ment, later writers have been less willing to accept metastasizing tumours as mesothelioma! McOaughcy (1958) was rather guarded on the subject mentioning metastases to lung and liver but regarding more distant spread as unusual though not in itself a reason for changing tht diagnosis. In other accounts (Churg. Rosen anil Moolten, 1965; Hourihane, 1964; Payling Wright and Heard, 1966) the infrequence of distant meta stases has been emphasized, so much so that whet these have been present the diagnosis has beet regarded as uncertain or even been rejected Evans (1966) summarizes a commonly held vie when he says `at the present stage of our know ledge of these tumours the diagnosis in such cases is open to question since the distinction from other widely metastasizing neoplasms is difficult'. Wide adoption of this view has meant that only' mesotheliomas with no or very limited metastases have been accepted as mesotheliomas in man) published accounts, perpetuating a misconception about the nature of the tumour. Originally we shared the view that these tumours rarely metastasize, but with increasing experience we have, like Clnirg and Selikoi (1968), come to accept metastases as of frequent occurrence. CLINICAL FEATURES ` The commonest presenting complaints wert, breathlessness and pain in the chest, and on first examination a large pleural effusion was oftet present. The most persistent symptom was pain, while dyspnoea was often relieved for several months after aspiration of pleural effusions. Tea common initial symptoms were cough and loss of weigh! (S patients), congestive cardiac failure (3 patients), and haemoptysis (1 patient). In 2? case record clubbing at in 16 cases, < in live cases AGE (FIG. 1-1 a boy aged Ithe ages of sought mci <74%) were i mean 604 I' 45 40 35 JO- t 25 l I 15 to 5 O O L Age (y no, 14. /T sex Severn men, reflect Merseyside hazard. In tact could occupation exes equa significant. DURATION i not possible symptoms, inadequate imposed <bronchitis patients thi illness from anl is sh<> per cent ot vivttl two \ and 8% O' Diffuse malignant pleural mesothelioma and asbestos exposure 17 resent in one in a : was a the right iVpes of .L-.es. but >did so, ied this - in the mcreased e of the the pro be retro.ah non- case records there was comment about finger clubbing at the first examination. It was absent in 16 cases, doubtful in six, and definitely present in five cases. ace (fig. 14) Apart from one mesothelioma in a boy aged 16 years, all the patients were between the ages of 40 and 82 years at the time they sought medical advice. Thirty-eight patients (74%) were between the ages of 50 and 70 years, mean 6010 years SD. ,-xtcnsive mt comHing to tioliomas. < on the .'.nd liver unusual, ging the oxen and ig Wright .nt metahat when has been rejected, held view ur know-uch cases lion from difficult', that only netastases in many (inception out these increasing .1 Selikoff frequent ints were | id on first* Mas often | was pain, I >r several ( aonx. Less { i and loss | ac failurel nt). In 27 I Survival time (years) no. 15. Total duration of illness of 36 mesothelioma patients from first symptoms to death. Age (years) fig. 14. Age distribution o f 52 mesothelioma patients. SEX Seventy-seven per cent of the patients were men, reflecting the fact that asbestos exposure on Merseyside is predominantly a male occupational hazard. In the few cases where no asbestos con tact could be established from the patients' occupations the disease appeared to affect both sexes equally, but too few were involved to be significant. duration o f il l n e s s In some patients it was not possible to date accurately the first onset of symptoms, either because the records were inadequate or because the symptoms were super imposed on those from pre-existing chronic bronchitis and emphysema. However, in 36 patients this was possible, and so the duration of illness from first symptoms until death was known, and is shown graphically in Fig. 15. Fifty-three per cent of patients survived one year, 33% sur vived two years, 17% were alive after three years, and 8% (three cases) lived over five years. These figures show less favourable results than might be expected in uncomplicated cases without sur gical treatment, because two died from unrelated coronary thrombosis, one from a bleeding duo denal ulcer, and two from postoperative pulmonary embolism. There was no evidence that radical surgery or chemotherapy affected the prognosis favourably. The relatively long survival with only slow deterioration of many mesothelioma patients has received little attention, except from Le Roux (1962). Hinson (1965) even says that death usually occurs within a year of the first symptoms. 13ccau.sc of this slow deterioration we find that clinicians caring for mesothelioma patients, but more familiar with cases of inoperable lung cancer, often ask for the surgical histology to be reviewed two or three years after the original report. A direct comparison of live survivals of both inoperable lung cancer and pleural mesothelioma, from the time of histologica! diagnosis, has been made to illustrate this difference in survival (Fig. 16). The lung cancer patients were a series of 419 cases diagnosed by bronchial biopsy at Broadgreen Hospital during 1950-5 but considered to be inoperable (Whitwell, 1961). The meso theliomas are the 30 in the present series diag nosed from operation specimens. The graph underestimates the mesothelioma survival because of the relatively higher early mortality after pleurcctomy or decortication compared with bronchial biopsy, and because the mesothelioma group 18 /'. W h it well a n d R a ch el M . R a w cliffe f i g . 16. Comparison o f survival rates o f bronchial carci noma and pleural mesotheliomas from histological diagno sis to death includes live living patients who have been listed according to their present survivals of between 6 and 36 months. At six months 20 ;;, of lung cancer patients and 66;,, of mesothelioma patients remained alive, while at 12 months only 4% of lung cancer patients remained alive but there were still 43",, of the mesothelioma patients. All the lung cancer patients had died by 18 months, but 26",', of the mesothelioma patients survived, and one of these survived over four years. This finding disagrees with that of Elmes (1966) who found that few patients survived more than a year from the time that the diagnosis had been established. cig a r e t t e s m o k i n g It h a s b e e n s u g g e s t e d t h a t a further factor in addition to asbestos is necessary for the developm ent of mesothelioma, so the pos sible association of cigarette sm oking and asbestos with m esotheliom a has been considered in the present series. The smoking habits of 28 patients had been recorded and included 10 non-smokers, eight who had smoked up to 10 cigarettes a day, and 10 who smoked more than this. The incubation period from first exposure to asbestos until development of tumour was no shorter in the smokers than in the non-smokers, neither was the duration of asbestos exposure any shorter in the smokers. INDUSTRIAL EXPOSURE TO ASBESTOS male p atien ts Thirty-five of the 40 patients with mesotheliomas had experienced significant exposure to asbestos dust. In 28 cases this had been incurred through their trades in shipbuilding and repairing firms, and in three further cases; with numerous asbestos bodies in the lungs the' patients had spent a large part of their working lives in dock-side occupations. Only four patients; in the series had been exposed to asbestos while following non-maritime occupations--two laggers, one asbestos factory worker, and one brake-lining factory employee. The trades of the affected shipyard workers are; given in the Table, showing that at least a dozer different occupations are at risk. Many of these; workers had a vague knowledge of the hazards to- health from asbestos but some denied any contact; with asbestos though their lungs contained numerous asbestos bodies. It seems that the exact; nature of their work was much less important' than the fact that they all worked in shipyards.! Elmes and his colleagues (1965) have graded the) work of shipyard employees with mesotheliomas": into that with heavy, light, and no asbestos risk, but Harries (1968) has shown from his work in) Royal Naval dockyards that the risks of meso-' thelioma are not confined to those workmen who" actually handle asbestos, and he mentions theij trades of men with mesothelioma as boilermaker, titter, labourer, shipwright, and welder. r l: i TABLE j SHIPYARD TRADES 01- 2 PATIENTS WITH PLEURAL MESOTHELIOMA Laisscr ........................................................ Jomor Boiler maker Boiler sealer Sheet-metal worker hngineer Lnginoer's litter Plater Welder apprentice .. Plumber Engineer's labourer Labourer Lorry driver No. of Patients 4 i 2 ) 2 2 > ) 1 ! I I MAI E PATIENTS thelioma were eigl industrial asbestos others two had no asbestos bodies in and no asbestos bo, no hazard remains ' Of the eight p mesotheliomas, thi, three necropsy mi numerous asbestos arc still alive. Two of the eigl lioma patients had asbestos factories I , side. The other six. have suffered induh.ul worked in van These patients h.n bags and sacks ai hessian sacks whs asbestos. Owen 0 " these cases, drawi, occupational risk f mesothelioma path have found is one New house and TL 1heir case 76, desc; was classified anu asbestos contact bn group. It is difllcu1 is either largely ct sacks for repair in more tree of asbest further cases from flic four non-inu In a housewife win village in the Isle hrewerv worker wi labelling plant, am h.ul worked on gui of these showed lungs but no mm. occasionally in the The five cases with no occupational history of! asbestos exposure were a grocer, a chlorine plank operator, a leather worker, an electrician, and a schoolboy. In no case was there any evidence of; home or environmental exposure to asbestos. All! died without necropsy studies except the leather; worker, and no asbestos bodies were found in hit) lungs; Though some electricians are exposed to1 asbestos dust there was no history of asbestos exposure in the electrician with the mesothelioma. DURATION OF EXPO- shipyard, asbestos ! vtck repair factm j<riod of exposui had in their later v vivory posts and casbestos dust. The period of c patients is known from four path i; Plit' Diffuse malignant pleural mesothelioma anti asbestos exposure 19 had |i ling i ises the i King | ents : hile L ers, ning I. v'r a r e >/en hese I; Is to B n ta c t It ned exact Tant rds. I the nas risk, ik in E teso- it i who h k the > .iker, t i t RAL ;ory of e plant and a ; nee of ft' tos. All leather ;. d in his osed to ; .sbestos , iolioma. female patients Among the 12 cases of meso thelioma were eight with a definite history of industrial asbestos dust exposure, and of the others two had no known hazard but had scanty asbestos bodies in the lungs, one had no hazard and no asbestos bodies in the lungs, and one with no hazard remains alive. Of the eight patients with asbestos-induced mesotheliomas, three died without necropsy, in three necropsy studies showed moderate or numerous asbestos bodies in the lungs, and two are still alive. Two of the eight asbestos-induced mesothe lioma patients had been exposed to asbestos in asbestos factories before they moved to Mersey side. The other six, who are the only women to have suffered industrial exposure on Merseyside, had worked in various bag and sack repair works. These patients had been sorters or repairers of bags and sacks and they dealt with dusty used hessian sacks which had sometimes contained asbestos. Owen (19b41 described the first two of these cases, drawing attention to this unusual occupational risk from asbestos: the only other mesothelioma patient of similar occupation we have found is one in the large series described by Newhouse and Thompson (1965) from London. Their case 76, described as a remaker of old sacks, was classified among those with no history of asbestos contact but clearly belongs in the same group. It is difficult to believe that this industry is either largely confined to Merseyside or that sacks for repair in other parts of the world are more free of asbestos dust, so it is likely that many further cases from this occupation will be found. The four non-industrial mesotheliomas occurred in a housewife who had alsvay.s lived in a small village in the Isle of Man, a grocery assistant, a brewery worker who only worked in the bottlelabelling plant, and a football-pool checker who had worked on gun drilling during the war. Two of these showed scanty asbestos bodies in the lungs but no more than is likely to be found occasionally in the adult female population. duration o f ex posu re The to ta l tim e s p e n t in a shipyard, asbestos factory, brake-lining factory, or sack repair factory has been regarded as the period of exposure, even though some patients had in their la te r y e a rs b e e n p r o m o t e d t o s u p e r visory posts and escaped rather too late from the asbestos dust. The period of exposure to asbestos dust of 36 patients is known and shows a very wide range, from four patients with only three years' exposure to four patients with over 50 years' exposure. The four patients with short exposures were questioned closely about their other jobs but none seemed relevant, and although these patients had ceased to be exposed to asbestos over 20 years before developing their mesotheliomas, the three who died showed asbestos bodies in their lungs. The mean exposure period of the series is 28 + 16 years SO, rather shorter than the mean of 35 years found in the Belfast shipyards (Elmes et al.. 1965). There was no significant dillerence in dura tion of exposure between one hazardous trade and another, or between men and women. INTERVAL BETWEEN FIRST' ASHES LOS EXPOSURE AND f i r s t m e s o t h e l i o m a s y m p t o m s The date of onset of mesothelioma symptoms and the exact period of asbestos exposure were known in 33 patients and are shown for the shipyard and sackware workers (Fig. 17). The interval from first exposure to asbestos until first mesothelioma symptoms is a much less variable figure than that for total asbestos exposure. Apart from two short periods of 13 years and 19 years the interval was always over 25 years, and in I 1 patients was over 50 years. The mean value of this interval for the 33 cases was 42 ; 12 years SD, which is surprisingly close to the Belfast figure of 43 + 13 years SD for the slightly dillerent interval between first asbestos exposure and death (Elmes, 1966). As the minimum exposure period in the series was three years it is likely that the induction period of an asbestosinduced mesothelioma is about 40 years. Ll'NO CHANGES IN ASBESTOS-INDUCED PLEURAL m e so t h e l io m a s Lungs front 30 mesothelioma patients who had a history of asbestos exposure were examined at necropsy and asbestos bodies were found in all of them. However, as these bodies have been found recently in a high pro portion of urban male lungs and in a slightly lower proportion of urban female lungs, their presence does not itself imply excessive or indus trial exposure to asbestos. The most certain evidence of excessive asbestos exposure is the presence of asbestosis, and basal asbestosis was present in live cases (17%). These were three luggers, a shipyard l a b o u r e r , a n d a boilermaker. In 10 other cases asbestos bodies were so numerous that they were immediately seen, often in clumps, in routine sections. In 15 cases more prolonged search was required to line! asbestos bodies. f. 20 F. W hitwell and Rachel M . Rawcliffe 00 o2r o o o o o o 91 92 91 - r-J r**j -sj- u "> -JD lirms, one was worked in a sh I.ting juice Shi pbuildinq Industry \mm 1 '////,Mf. m ! 1 ('o>//: - wfW,'S3 Tuiimnrr L it T. ...... m 1 ii Boiler scaler Engineer's fitter Joiner Joiner No lung tissue + *h + + induced mesotl examined quai> cases, which dl tosix. Only one bodies per sme between 1929 . live cases sho TmJIMTf/-O mm VW'wiWii.m m Welder Plater Plumber No lunq tissue. Alive + + + + bodies per snn between 25 an xmcar. We thn lung juice asln i \m \ u bi rm rm lk m i \ i Wifi,3 1 -- 11 1 TTv) 1 W'\ 1 H i%>/]1 1 0 I I ['//%Wtl' 1 D M i: H Labourer Laqqer Boiler maker Sheet-metal worker Laqqer Laqqer Boiler maker Sheet-metal worker Oxyacetylene cutter Boiler scaler Engineer's labourer Apprentice Lorry driver Joiner + + + + Asbestosis + + Asbestosis Asbestosis No lunq tissue + No lunq tissue + + No lunq tissue. Alive + + + considerable ' from urban ac The distinct a pscudo-asbe nature of its parent in the l carbon in the resembling tl. found is a tab the 200 postm giecn Hospit bodies, but wi mg asbestos asbestos bodic the smears - asbestos bodiasbestos bodi cases. Sackware Industry s' niiNricm 0m m 'W'W-'S/'y.y'M i '] m i| I | | Sackware sorter Sackware repairer Sackware sorter Sackware sorter ++ No lunq tissue No lunq tissue + The argument tl-'Tioinas exi tumours app, 10 years in f growths with Miidy we cer' J L I..L1 Period of asbestosexposure , + In moderate numbers ,, , , I Asbestos bodies , , . , , theliomas as I Appearance of mesothelioma | -M-ln-arge numbers .ualilied acc I f ig , 17. Time relationship o f industrial exposure to asbestos with the develop flave non-ind- ment o f mesothelioma and the state o f the Innas. to adequate x industriai me t i .unc; s m e a r s In 1966-7 lung smears were number of asbestos bodies in these patients wat three neerop ( examined routinely at Broadgrcen Hospital from nearly always far less than is found after indus sarcomatous 200 consecutive necropsies on patients of botli trial exposure. Of the 48 positive smears, 41 (85%) theliomas. T : t ksenxoews novcaesresthoef apgneeuomfoc3o0nioyseiasr.s,Aesbxecslutodsingbodaineys ssehvoewnedpalteisesnttshasnhow12edbobdeitewseepner2s5maenadr, 5a0ndbodoniel)s ptiaicpilhliasrtyolotugny were present in 24% of the series, in 33% of men per smear. These seven may well have been staining in b and in 13% of women. The most strongly positive exposed to asbestos in industry, for they were all i.rncly seen group of men between 50 and 70 years of age, elderly men with few occupational notes, excepl which raises of whom 39% showed asbestos bodies. But the that two had worked as painters for industrial die diagnosi of pleural t r & Diffuse malignant pleural mesothelioma and asbestos exposure 21 i icnts was er indus4! (85%) \ and only ' '0 bodies ; :ve been were all [ s. except industrial firms, one was a wood machinist, and one had worked in a shipyard. Lung juice smears made on the asbestosinduced mesothelioma patienls' lungs have been examined quantitatively in the last 14 necropsy cases, which did not include any showing asbestosis. Only one case showed less than 11 asbestos bodies per smear, and he had been a boiler scaler between 1929 and 1932 and afterwards a grocer. Five cases showed between 12 and 24 asbestos bodies per smear, and in eight cases there were between 25 and over 1,000 asbestos bodies per smear. We think that when asbestosis is absent, lung juice asbestos body smear counts can be of considerable value in differentiating industrial from urban asbestos exposure. The distinction between an asbestos body and a pseudo-asbestos body is made easily from the nature of its central substance, which is trans parent in the true body and usually black due to carbon in the pseudo-body. The only structures resembling the true asbestos body likely to be found is a talc body, which js much shorter. Of the 200 postmortem smears examined at Broadgreen Hospital 12% showed pseudo-asbestos bodies, but whereas 25% of the smears contain ing asbestos bodies also contained pseudo asbestos bodies, these were present in only 8'% of the smears without asbestos bodies. Pseudo asbestos bodies were also sometimes found with asbestos bodies in smears from mesothelioma cases. AUTHENTICITY OF PRIMARY PLEURAL MESOTHELIOMA The argument as to whether primary pleural meso theliomas exist or are nearly always secondary tumours appears to have been settled in the last 10 years in favour of accepting them as primary growths with only a few abstentions. From our study we certainly accept asbestos-induced meso theliomas as primary neoplasms but make this qualified acceptance because only very rarely have non-industrial mesotheliomas been subjected to adequate study. Of our nine patients with nonindustrial mesotheliomas, all have died, but only three necropsies were carried out, confirming one sarcomatous and two tubulo-papillary meso theliomas. The six other cases were till tubulopapillary tumours, but three were atypical in that the histology showed faint positive muciearmine staining in the tumour cells and acini, features rarely seen in asbestos-induced mesotheliomas, which raises slight doubt about the exactness of the diagnosis, which was based on a small amount of pleural tissue Even if all these nine cases arc acceptable as mesotheliomas, the non-asbestos-induced meso thelioma remains a very rare tumour, which has appeared in our three hospitals about once in two years. In this country the most persistent opposer to the existence of the pleural mesothelioma has long been Willis (1952), who gives personal experience and quotes previously published papers on the subjectTthe latest being from 1935 and 1942) in which he finds valid reasons for non-acceptance of the diagnosis. It is interesting that this work occurred in the first half of the century and is really a criticism against the existence of a nonasbestos-induced mesothelioma. With the wide spread use of asbestos starting only at the begin ning of this century, and a mean tumour induction period of around 40 years, there can have been only very few asbestos-induced mesotheliomas in the world at that period. In his later work Willis (1967) criticized many of the writers quoted in the present paper, largely on the basis of in adequate documentation or illustration of cases, or lack of necropsy studies. He suggested that the diagnosis of mesothelioma should be made only after full postmortem examination, including the fine slicing of fixed lungs to search for small subpleural carcinomas. In the present series no line slicing was necessary to see subpleural tumours, but these were mtastass from the primary growths. If the majority of mesotheliomas were accepted as secondary carcinomas it would be impossible to explain their curious occupational incidence, their prolonged survival compared with accepted bronchial carcinomas, or their consistent macroscopic appearance, which is not typical of carcinoma. INCREASING INCIDENCE O f ASBESTOS-INDfC ED PLEURAL MESOTMEI IOMA Between 1955 and 1963 Owen (1964) found 12 histologically confirmed asbestos-induced meso theliomas, and in the same hospitals between 1964 and 1970 we have found 31 additional cases. The pathology records at Broadgreen Hospital show that two cases occurred between 1950 and 1954: these figures suggest to us that there is real and large increase in the incidence of this tumour. Even after working on this subject for a long time it is difficult to become adapted to the amaz ingly long induction period shown by some of the patients. Several of the shipyard workers examined in the last five years first inhaled asbestos when building ships which fought at the battle of Jut- ..ti.uiu iWwv*-iff**-*!-*"*"" -*< F. Whtwell and Rachel M . RawcH/Je land, while their younger mates with meso theliomas first contacted asbestos while building ships for the second world war. Our successors will, in about 40 years time, be seeing meso thelioma patients who first inhaled asbestos while building post-war nuclear submarines. Even if advice contained in the white paper 'Problems Arising from the Use of Asbestos* (1067) is adopted by industry without delay, it will be some time in the twenty-first century be fore the incidence of pleural mesothelioma ceases to rise. We wish to thank Dr. W. GJyn Owen for the use of his valuable notes, not only on the cases in the scries that he published but also on a few of the later cases. Wc are grateful to o u r colleagues at Broadgreen, Aintree, and St. Catherine's Hospitals for making available the records of their patients, in particular Dr. D. L, Caldwell, of St. Catherine's Hos pital, under whom were admitted most of the ship yard patients, and Dr. P. J. Taylor, of Aintree Hos pital, who carried out several of the postmortem examinations. We arc also indebted to Mr. A. Robertson and Miss J. Scott for histological prepara tions and for performing numerous asbestos body counts on lung .smears. REFERENCES Ashcroft, T. (196S). Asbestos bodies in routine necropsies ,>n Tyneside, hi!, tried. J., 1. 614. C flag . J., Rosen, S. H., and Moot ten, S. (1965). Histological characteristics of mesothelioma associated with asbes tos. Ann. ,Y. Acad. .Sr/., 132, 614. ------and Sclikotf, I. J. (19681. Geographic pathology of pleural mesothelioma. In The Lung. International Academy o f Pathology Monograph, p. 284. Edited by A. A. Eiebov. and D. E. Smith. Williams and Williams, Baltimore. Elmes, P. C. (196(>). The epidemiology and clinical features of asbestosis and related diseases. Postgrad. and. J ., 42, 623. McCaughey. W. T. D, and Wade, O. 1.. (1965). Diffuse mesothelioma of the pleura and asbestos. Brit. nu-d. ./., I, 350. Hntieknap, J. B., and Smither, W. J. (1964). Peritoneal tumours in asbestosis. Brit. J. indto.tr. Met/., 21, 20. Evans, R. Winston (1966). Histological Appearances cj Tumours, 2nd ed., p. MS. Livingstone, Edinburgh and London. Godwin, M. C. (1957). Diffuse mesotheliomas, with com ment on their relation to localised fibrous meso theliomas. Cancer (Philad.), 10, 298. Harries, P. G. Asbestos hazards in Naval Dock yards. Ann. theup. Hyp., II, 135. Heard, IE I:. (1966/. Asbestosis. In Recent Advances ii Pathology* 8th ed. Edited by t \ V. Harrison, p. 366, Churchill, London. Hinson, K. F. W. (1965). Cancer of the lungs and other' diseases after exposure to asbestos dust, Brit. J. Dh. Chest, 59, 121. Hourihane, D. O 'H. 0 964). The pathology of mesotheliomata and an analysis of their association with asbestft exposure. `Thorax, 19, 268. Le Roux, 1L T. (1462). Pleural tumours. Thorax. 17, III. ; MeC'aughev, W. T. L. (19581. Primarv tumours of the pleurt J. Path. Had.. 76, 517. Newhouse, M. L., and Thompson. M. (1965). Mesidhcliomi of pleura and peritoneum following exposure to asbestos in the London area, ii'ii. J. indu.ur. MeiL 22, 261. . Owen. W. Glyn (1964). Diifu.se mesothelioma and exposure to asbestos dust in the Merscv-nde area. Brit, tried. J, 2,214. Problems Arising from the Use o f Asbestos. (1967). II.M.S.O, London. Roberts, G. Hetin (1967). Asbestos bodies in lungs at necropsy. J. din. Path., 20, 570. Wagner, J. C\, Monday, D. E-, and Hariiigton, J. S. (1962) Histochemical demonstration of hyaluronic acid it pleural mesotheliomas. J. Path. Bad.. 84, 73. 1 Sleggs, C. A., and Marchand, I*. (1960). DilTust pleural mesothelioma and asbestos exposure in tlK North Western Cape Pro\ince. Brit J. industr. Mei 17, 260. Whitwell, F. (1961). The histopalhology of lung cancer ii Liverpool: the specificity of the histological cell type, of lung cancer. Brit. J. Cancer, 15, 440. Willis, R. A. (1952). The Spread o f Tumours in the Htami Body, 2nd eL. p. 55. Butterworths, London. ----- (1967). Path<>l(>py o f Tumours, 4th ed., p. 181. Butter- worths, London. Wright, Payling G , and Heard, B. E. (966). In S/stemk Pathology, by (5. Payling Wright, and W. St. C Symmers, \ol. 1, p. 426. Longmans, London. j Li R. The M- The lun; broncho broncho Ihc rcl.it lo be in. by tunn DeK. chronic In thi picdicin a pnow. may K I Unjj SCai" demons! ra: blood flov. I'*>4 ; Tan been used pahents vs Maxficld. I raser, am vary in si/i visible on vion of an arc comnn This pa] ihe bronc! <n patient between !' value of i! The paticn. rho bronci' 11 dia. l cvcloapbylivohredl resected ti l Mach pa1 *" l.ibclK Intrsivcnoi. f Picker M colour scar i'trtrfrt au uc Wdifviis I