Downloaded from thorax.bmj.com on November 21, 2011 - Published by group.bmj.com
860 Thorax 2000;55:860–863
Malignant vascular tumours of the pleura in
“asbestos” workers and endothelial diVerentiation
in malignant mesothelioma
R L Attanoos, S K Suvarna, E Rhead, M Stephens, T J Locke, M N Sheppard,
F D Pooley, A R Gibbs
Abstract made from this small series but further
Background—Three cases of diVuse ma- investigation is warranted.
lignant vascular tumours of the pleura are (Thorax 2000;55:860–863)
described which mimicked malignant
mesothelioma clinically and pathologi- Keywords: pleura; epithelioid haemangioendothelioma;
asbestos; mesothelioma
cally (so called “pseudomesothelioma”).
All had occupational histories of exposure
to asbestos. The relationship of these There are well established associations be-
tumours to mesothelioma and asbestos tween exposure to asbestos and several pleural
exposure is discussed. conditions including hyaline plaques, diVuse
Methods—To examine the histogenetic fibrosis, and malignant mesothelioma1–3 but no
relationship between mesothelioma and association with vascular tumours of the pleura
these three tumours an immunohisto- has been made. In this report three cases of
chemical analysis of vascular marker primary pleural epithelioid haemangioendo-
Department of
Histopathology,
(CD31, CD34, and Von Willebrand factor) thelioma are described which have occurred in
Llandough Hospital, expression was undertaken in 92 cases of subjects with a history of occupational expo-
CardiV, UK pleural mesothelioma, in addition to these sure to asbestos who have had a full electron
R L Attanoos three tumours. Electron microscopic fibre microscopic fibre analysis of lung tissues. The
A R Gibbs analysis of lung tissue was performed on diVerential diagnosis between these tumours
each of the three cases to assess asbestos and malignant mesothelioma is discussed par-
Department of
Histopathology,
fibre content. ticularly with regard to their immunohisto-
Northern General Results—DiVuse pleural epithelioid haem- chemical profiles. The medicolegal implica-
Hospital, SheYeld, UK angioendotheliomas may closely resemble tions of the diagnosis of pleural epithelioid
S K Suvarna malignant mesothelioma clinically and haemangioendothelioma are considered.
pathologically but, of the 92 pleural meso-
Department of theliomas tested, none showed expression
Histopathology, North
StaVordshire General
of CD31, CD34, and Von Willebrand Case reports
Hospital, Stoke on factor. Although all three cases had The three cases presented were each referred
Trent, UK claimed exposure to asbestos, ferruginous for pathological review during the process of
E Rhead bodies typical of asbestos were only seen medicolegal compensation claims for asbestos
M Stephens by light microscopy in case 2, and only in related disease. The clinical data and necro-
this subject was the asbestos fibre content scopic information from the three patients with
Department of
Histopathology, Royal
raised in comparison with the range seen primary pleural epithelioid haemangio-
Brompton Hospital, in a non-exposed background population. endothelioma are shown in table 1. The occu-
London, UK The latent period in the pleural epithelioid pational asbestos exposure is given in table 2.
M N Sheppard haemangioendotheliomas ranged from 18 Necroscopic examination showed that the
to 60 years. pleura was the sole or major site of disease. The
Department of Conclusions—Endothelial diVerentiation pathological diagnosis was independently
Surgery, Northern
General Hospital,
does not appear to occur in mesothelioma made by at least two experienced pulmonary
SheYeld, UK and therefore should be clearly separated histopathologists after consideration of haema-
T J Locke from it. No definite association between toxylin and eosin stained slides and a wide
pleural epithelioid haemangioendothe- panel of immunohistochemical markers. In all
Division of Materials lioma and exposure to asbestos can be cases ultrastructural studies were performed.
and Minerals,
University of Wales,
CardiV, UK Table 1 Clinical features of primary pleural angiosarcomas (epithelioid haemangioendothelioma)
F D Pooley
Course
Correspondence to: Case Sex Age (years) Clinical Radiology (months) Necroscopic findings
Dr R L Attanoos, 1 M 59 Pleuritic pain, EVusion, diVuse 7 DiVuse pleural tumour,
Department of dyspnoea pleural thickening, hyaline pleural plaques,
Histopathology, Llandough plaques (CXR, CT) ferruginous bodies in lung
Hospital, CardiV CF64 2XX, 2 M 73 Chest infection, EVusion, pleural 9 DiVuse pleural tumour,
UK dyspnoea fibrosis,pleural mass, pleural fibrosis, classical
plaque (CXR,CT) asbestos bodies in lung
Received 11 November 1999 3 M 33 Cough, pleuritic EVusion, parietal 8 DiVuse pleural neoplasm
Returned to authors pain, dyspnoea pleural thickening, with local invasion of
20 March 2000 diaphragm and mediastinum and
Revised version received pericardial spread diaphragm. Contralateral
11 July 2000 (CXR, CT) plaque. Ferruginous bodies,
Accepted for publication many coarse “welding type”
12 July 2000
www.thoraxjnl.com
Downloaded from thorax.bmj.com on November 21, 2011 - Published by group.bmj.com
Serosal vascular tumours and asbestos 861
Table 2 Occupational details, tumour latency and cumulative asbestos exposure dothelial diVerentiation in cases 1 and 2 (fig 2).
The diagnosis in both cases was of primary
First exposure and Latency
Case Occupation duration (years) Intensity pleural epithelioid heamangioendothelioma. In
case 3 the tissue was suboptimal for ultrastruc-
1 Labourer 1956–8 (2 years) 42 Light, unknown tural analysis. Four independent pathological
frequency
2 Shipyard joiner 1938–43 (5 years 60 Heavy, intermittent opinions were sought and the overall favoured
3 Shipyard welder 1975–6 (18 months) 17 Heavy, intermittent diagnosis was of primary pleural angiosarcoma
(epithelioid heamangioendothelioma).
Methods
Mineral fibre analysis was undertaken on
representative lung tissue blocks. Lung tissue
samples were digested in 40% potassium
hydroxide, washed, and ashed at 350°C in oxy-
gen. The abstract was suspended in distilled
water and aliquots were filtered on nucleopore
filters. These were then carbon coated and the
filters dissolved in chloroform. The carbon
films were mounted onto gold electron micro-
scope support grids for transmission electron
microscopy. Fibres were counted and typed by
Figure 1 Epithelioid haemangioendothelioma with spindle an energy dispersive x ray analysis technique.4
and vacuolated cell forms showing neovascular lumina with The potential role of immunohistochemistry
erythrocytes (arrow). Stain: H & E, ×400 magnification. in the diagnosis and distinction of malignant
vascular tumours of the pleura from malignant
mesothelioma prompted an analysis of endo-
thelial marker expression in malignant meso-
thelioma. ParaYn embedded blocks from 92
cases of malignant mesothelioma (84 Lland-
ough Hospital, Penarth/8 Northern General,
SheYeld) were retrieved from the files. Stand-
ard sections were cut and stained with haema-
toxylin and eosin for morphological assess-
ment. Each case was chosen on account of its
typical light microscopic appearance for malig-
nant mesothelioma and supportive immuno-
phenotype by use of a broad antibody panel.
An immunohistochemical study was per-
formed using an avidin/biotin/peroxidase com-
plex (ABC method) with antibodies to von
Willebrand factor (Dako, 1 in 200 dilution),
CD34 (Europath, 1 in 200 dilution), and
Figure 2 Epithelioid haemangioendothelioma (case 1) CD31 (Dako, 1 in 20 dilution). Suitable posi-
containing intracytoplasmic Weibal-Palade organelle
(arrow). EM, ×57 000 magnification. tive and negative control tissues were included.
In each case the parietal pleura was diVusely
Results
replaced by tumour composed of epithelioid
The results of the mineral analysis are shown in
and vacuolated cells forming small nests and
table 3. The putative latent periods of the three
cords with a minor spindle cell component.
cases ranged from 18 to 60 years, with cumula-
The tumour cells showed a moderate degree of
tive exposures to asbestos of 2–17 years. In
cytonuclear pleomorphism and intracellular
cases 1 and 3 the intensity of asbestos exposure
neovascular lumina containing erythrocytes
was light. In both cases ferruginous bodies
were present (fig 1). Immunohistochemistry
were present but they were typical of asbestos
revealed that the tumour cells were positive for
only in case 2. In this case there was intermit-
vimentin and with at least two of the three
tent heavy exposure to asbestos and lung fibre
endothelial markers (CD 31, CD34, von
burdens reflect the occupational history. Cases
Willebrand factor). In case 3 this additionally
1 and 3 do not show a clear cut increase in
included the vascular marker thrombo-
asbestos fibre levels.
modulin. The identification of intracytoplas-
The results of the immunohistochemical
mic Weibal-Palade organelles confirmed en-
study of vascular marker expression in malig-
Table 3 Electron microscopic mineral analysis on lung tissue (fibres/g dry lung tissue × nant mesothelioma shows that there was no
106) immunoreactivity for CD31, CD34, or von
Willebrand factor in any of the subtypes of the
Case Total Amosite Crocidolite Chrysotile Mullite Others
92 malignant mesotheliomas tested.
1 27.2 ND ND ND 8.8 18.4 (0.5
tremolite)
2 51.7 12.3 29.5 ND 7.5 2.4 Discussion
3 6.9 0.4 ND 0.2 5.8 0.5 Epithelioid haemangioendothelioma is an un-
In case 1 non-fibrous mineral analysis revealed that 90% of particulate matter was mica. common low grade sarcoma of endothelial
ND = not detected. origin recognised in a variety of tissue sites
www.thoraxjnl.com
Downloaded from thorax.bmj.com on November 21, 2011 - Published by group.bmj.com
862 Attanoos, Suvarna, Rhead, et al
including liver, soft tissue, lung, bone, and asbestos, (3) an appropriate latent period, and
skin.5 It has been described as originating from (4) an absence of confounding aetiological fac-
the pleura6 and peritoneum.7 tors such as irradiation or erionite mineral fibre
In the pleura the main diVerential diagnosis exposure.
is that of diVuse malignant mesothelioma8 and The diagnosis of these three cases has been
clinicopathological distinction can be problem- discussed. The evidence of asbestos exposure
atical. Firstly, a diVuse serosal growth pattern is can come from various sources including occu-
a characteristic feature of malignant meso- pational history, light microscopic identifica-
thelioma but this “mesotheliomatous” pattern tion of asbestos bodies in lung tissue sections,
is not specific and has been recognised in some and mineral fibre analysis of tissue digests. The
non-mesothelial tumours, most notably carci- reliability of occupational history is strongly
nomas and, very rarely, sarcomas including dependent on “recall” which varies between
epithelioid haemangioendothelioma. Secondly, occupations and is subject to bias whereby
on morphological grounds, epithelioid haem- individuals can underestimate or overestimate
angioendothelioma may exhibit a biphasic pat- their exposure to asbestos. Electron micro-
tern with epithelioid and spindle areas similar scopic mineral fibre analysis allows for the
to biphasic malignant mesothelioma. accurate determination of asbestos fibre bur-
The accurate distinction of the two tumours den and fibre type at death. For case 2 all
has important medicolegal implications as only parameters indicated an increased crocidolite
malignant mesothelioma is currently recog- asbestos fibre count consistent with an occupa-
nised as an asbestos related neoplasm for com- tional exposure. However, for cases 1 and 3 the
pensation purposes. The description of three mineralogical results were equivocal. Case 1
primary pleural angiosarcomas in individuals worked most of his life as an underground coal
exposed to asbestos raises two important ques- miner but for a period of about two years (prior
tions. Firstly, do mesotheliomas have the to mining) worked as a labourer demolishing
capacity to undergo vascular diVerentiation sheds said to contain asbestos roofing materi-
and, secondly, if not, are pleural angiosarcomas als. Light microscopy showed occasional ferru-
aetiologically linked with asbestos? ginous bodies but no “classical” asbestos bod-
Divergent diVerentiation of mesothelium has ies. Mineral fibre analysis showed a low level of
been described and cases of malignant meso- tremolite, a contaminant amphibole of chrys-
thelioma showing leiomyoid, chondroid, os- otile (white) asbestos ore. It is possible that the
seous, and liposarcomatous patterns have been major asbestos exposure in this case was to
reported.9 Bona fide cases of malignant meso- chrysotile. However, it is well known that the
thelioma showing angiosarcomatous diVeren- biopersistence of chrysotile is low and, with
tiation have not been described. clearance over a 32 year period, the fibre may
In order to address the issue of vascular dif- have been undetectable by mineral analysis at
ferentiation in malignant mesothelioma an death. A further interesting finding in this case
immunohistochemical analysis of 92 mesothe- was that 90% of the non-fibrous particulate
lioma was undertaken using the three most matter was mica. The mica consisted of very
sensitive and specific vascular markers large plates and was not typical of that usually
(CD31,10 CD34,11 and von Willebrand seen in association with coal mining. Mica has
factor12). In one recent study two of 16 been associated with ferruginous body forma-
malignant mesotheliomas and five of 84 mixed tion (which are sometimes mistaken for
carcinomas showed weak non-specific cyto- “classical” asbestos bodies) and pleural
plasmic CD31 staining.13 In a further series14 plaques,15 and an awareness of this is important
no CD31 expression was seen in any of nine to prevent a false attribution to asbestos,
malignant mesotheliomas tested. The expres- particularly in the presence of a diVuse pleural
sion of von Willebrand factor and CD34 has tumour. There is no known association of mica
not been reported in malignant mesothelioma with malignant mesothelioma. In summary,
and the results of this study support previous only the occupational history indicated expo-
investigations.8 The present study indicates no sure to asbestos. In case 3 there was a positive
definite immunohistochemical evidence of vas- occupational history of exposure to asbestos as
cular diVerentiation in any of the cohort of 92 a shipyard welder but the chrysotile and
malignant mesotheliomas studied and we con- amosite asbestos fibre levels detected were not
clude that vascular marker expression (and clearly elevated above those identified in the
endothelial diVerentiation) does not occur or is background control population. Many ferrugi-
a very infrequent finding in malignant meso- nous bodies and iron coated platy material
thelioma. The use of a combination of the vas- were present, as frequently occurs in welders.16
cular markers CD31 and CD34 would there- An appropriate latent period (from initial
fore allow eVective distinction of malignant exposure to asbestos to death of the patient) is
mesothelioma from epithelioid angiosarcoma. an important factor in the ascription of an
Considering the aetiological role of asbestos, asbestos induced neoplasm. In a study of 1690
in the largest published series of 14 malignant cases of malignant mesotheliomas the latent
vascular tumours involving the serous period was in excess of 15 years in 99% of
membranes8 only one case of epithelioid cases.17 In comparison, in a series of occupa-
angiosarcoma had a prior history of asbestos tional related (hepatic) angiosarcomas follow-
exposure. In order to ascribe mesothelioma to ing exposure to vinyl chloride monomer18 the
exposure to asbestos certain conditions must latent period was 8–32 years with a mean
be met including (1) accurate diagnosis, (2) latency in excess of 20 years. In this series of
reliable evidence of exposure to amphibole pleural epithelioid haemangioendothelioma
www.thoraxjnl.com
Downloaded from thorax.bmj.com on November 21, 2011 - Published by group.bmj.com
Serosal vascular tumours and asbestos 863
the tumour latency period was similar to that bestos and, if so, to determine whether there
seen in asbestos related malignant mesothelio- are similar dose responses and latency periods
mas and ranged from 18 to 60 years. It is of as in malignant mesothelioma.
course recognised that, whilst the pathogenic
mechanisms of vinyl chloride and asbestos in 1 Churg A. Asbestos fibres and pleural plaques in a general
neoplastic transformation may be considerably autopsy population. Am J Pathol 1982;109:88–96.
2 Stephens M, Gibbs AR, Pooley FD, et al. Asbestos induced
diVerent, the studies do highlight the fact that diVuse pleural fibrosis: pathology and mineralogy. Thorax
long latent periods exist between initial expo- 1987;42:583–8.
3 Wagner JC, Sleggs CA, Marchand P. DiVuse pleural
sure to the carcinogen and death of the patient mesotheliomas and asbestos exposure in a north western
in the two tumour sites (liver and pleura). Cape Province. Br J Ind Med 1960;17:260–71.
4 Pooley FD, Ransome DL. Comparison of the results of
With respect to confounding factors, hepatic asbestos fibre dust counts in lung tissue obtained by
angiosarcomas have been aetiologically associ- analytical electron microscopy and light microscopy. J Clin
Pathol 1986;39:313–7.
ated with radiotherapy, anabolic steroids, vinyl 5 Weiss SW, Ishak KG, Dail DH, et al. Epithelioid
chloride monomer, and thorotrast. None of the haemangioendothelioma and related lesions. Semin Diagn
Pathol 1985;3:270–80.
patients in this series was known to have any of 6 Yousem SA, Hochholzer L. Unusual thoracic manifestations
these risk factors. Because of their rarity, no of epithelioid haemangioendothelioma. Arch Pathol Lab
Med 1987;111:459–63.
risk factors for pleural angiosarcomas have 7 Attanoos RL, Dallimore NS, Gibbs AR. Epithelioid
been clearly defined. haemangioendothelioma of the peritoneum: an unusual
mimic of peritoneal mesothelioma. Histopathology 1997;30:
In summary, three cases of primary epithe- 375–7.
lioid haemangioendothelioma of the pleura in 8 Lin BT-Y, Colby T, Gown AM, et al. Malignant vascular
tumours of the serous membranes mimicking mesothe-
patients who claimed prior exposure to asbes- lioma. Am J Surg Pathol 1996;20:1431–9.
tos are presented. The results of the immuno- 9 Donna A, Betta PG. Mesodermomas: a new embryological
approach to primary tumours of coelomic surfaces.
histochemical study do not support the view Histopathology 1981;5:31–44.
that endothelial diVerentiation occurs in malig- 10 Parums DV, Cordell JL, Micklem K, et al. JC 70: a new
monoclonal antibody that detects endothelial associated
nant mesotheliomas. We advocate the intro- antigen on routinely processed tissue sections. J Clin Pathol
duction of endothelial markers such as CD31 1990;45:752–7.
11 Van de Rijn M, Rouse RV. CD34. A review. Appl
and CD34 into the routine immunohisto- Immunohistochem 1994;2:71–80.
chemical panel used in the identification of 12 Guarda LA, Ordonez NG, Smith JI, et al. Immunoperoxi-
dase localisation of factor VIII in angiosarcomas. Arch
malignant mesothelioma so that eVective Pathol Lab Med 1982;106:515–6.
distinction from epithelioid haemangio- 13 Miettinen M, Lindenmayer AE, Chaubal A, et al. Endothe-
lial cell markers CD31, CD34, and BNH9 antibody to H-
endothelioma can be made. With regard to the and Y-antigens: evaluation of their specificity and sensitiv-
aetiology (that is, asbestos), it is recognised that ity in the diagnosis of vascular tumours and comparison
with von Willebrand factor. Mod Pathol 1994;7:82–90.
the three cases were identified during pursuit of 14 De Young BR, Frierson HF, Ly MN, et al. CD31 immuno-
medicolegal compensation claims for asbestos reactivity in carcinoma and mesotheliomas. Am J Clin
Pathol 1998;110:374–7.
related disease and therefore demonstrate 15 Davies D, Cotton R. Mica pneumoconiosis. Br J Ind Med
referral bias. However, further multicentre 1983;40:22–7.
16 Vallyathan V, Bergeron W, Robichaux P, et al. Pulmonary
case-control studies with mineral fibre analysis fibrosis in an aluminium arc welder. Chest 1982;81:372–4.
should be prompted by the findings of this 17 Lanphear BP, Buncher CR. Latent period of malignant
mesothelioma of occupational origin. J Occup Med 1992;7:
series to determine whether there is a genuine 718–22.
and consistent association between epithelioid 18 Evans DMD, Jones-Williams W, Kung ITM. Angiosarcoma
and hepatocellular carcinoma in vinyl chloride workers.
haemangioendothelioma and exposure to as- Histopathology 1983;7:377–88.
www.thoraxjnl.com
Downloaded from thorax.bmj.com on November 21, 2011 - Published by group.bmj.com
Malignant vascular tumours of the pleura in
''asbestos'' workers and endothelial
differentiation in malignant mesothelioma
R L Attanoos, S K Suvarna, E Rhead, et al.
Thorax 2000 55: 860-863
doi: 10.1136/thorax.55.10.860
Updated information and services can be found at:
http://thorax.bmj.com/content/55/10/860.full.html
These include:
References This article cites 17 articles, 3 of which can be accessed free at:
http://thorax.bmj.com/content/55/10/860.full.html#ref-list-1
Article cited in:
http://thorax.bmj.com/content/55/10/860.full.html#related-urls
Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.
Topic Articles on similar topics can be found in the following collections
Collections
Respiratory cancer (45 articles)
Environmental issues (155 articles)
Occupational and environmental medicine (75 articles)
Notes
To request permissions go to:
http://group.bmj.com/group/rights-licensing/permissions
To order reprints go to:
http://journals.bmj.com/cgi/reprintform
To subscribe to BMJ go to:
http://group.bmj.com/subscribe/