Document YkJopEk09xjJgGqwBj82qDrn

PANEE SILPANANTA, MD FERNANDO F. ILLESCAS, MD HUNTINGTON SHELDON, MD Thorotrast (colloidal thorium dioxide) has been known for 30 years to be a cause of malignant disease. Tumours have been reported in nasal sinuses, bones and liver, as well as other sites.'-7 This communication reports the discovery of two histologically distinct types of malignant tumour of the liver (angiosarcoma and hepatocellular carcinoma), as well as carcinoma of the lung and tubular adenomas of the kidneys, in a man who had undergone angiography with Thorotrast 38 years previously. Case report Clinical course A 74-year-old man was admitted to hospital because of increasing weakness, loss of appetite and progressive swelling of the abdomen for 3 months. Two weeks prior to admission he had begun to experience drowsiness and abdominal pain. His urine became darker, and an episode of diarrhea was reported. Thirty-eight years earlier, while in hospital for investigation of progressive failure of vision in his right eye, he had undergone bilateral internal carotid arteriography, with Thorotrast (total dose 36 ml) as the contrast medium. This showed a calcified lesion interpreted as an aneurysm of the right internal carotid artery. Two-stage ligation of the artery was performed. He had been in hospital several times for various medical and surgical problems. Fif- teen years before the current admission he had given up cigarettes after 20 years of smoking two packs a day. He denied excessive intake of alcohol. He was icteric and lethargic. His blood pressure was 150/70 mm Hg, and his temperature and pulse rate were normal. His chest was dull to percussion; wheezes and rhonchi were detected at the base of both lungs. Ascites and hepatomegaly were present. Neurologic examination yielded no abnormalities except for blindness of his right eye. The following abnormal laboratory findings were of note: blood erythrocyte count 2.9 X 1012/1, hemo- globin level 11.5 g/dl, hematocrit 32.7% and glucose level 143 mg/dl (7.9 mmol/l), total serum bilirubin level 3.8 g/dl (65 gmol/l) and serum levels of lactate dehydrogenase, glu- tamic oxaloacetic transaminase and alkaline phosphatase 416, 133 and 624 lU/l respectively. Tests for antinuclear, antimitochondrial and anti-smooth-muscle antibodies, a-fetoprotein (AFP) and hepatitis B surface antigen (HB5Ag), the last by radioimmunoassay, gave negative results. A chest roentgenogram demon- strated hyperinflation and an ill defined opacity in the lower lobe of the left lung that had not changed in appearance in 4 years. A plain film of the abdomen (Fig. 1) showed fine metallic densities outlining an irregularly shaped liver and a small spleen. Similar speckles were present in the hilar nodes of the spleen and the para-aortic nodes. From the departments of pathology and radtology, Royal Victoria Hospital and McGtll University school of medicine, Montreal Reprint requests to: Dr. Huntington Sheldon, Lyman Duff Medtcal Sciences Building, Department of pathology, McGill University, 3775 University St., Montreal, PQ H3A 2B4 FIG. 1-Plain film of abdomen, showing fine metallic densities, characteristic of Thorotrast deposition, in irregularly shaped liver, small spleen and lymph nodes. CAN MED ASSOC J, VOL. 128, FEBRUARY 1, 1983 289 The distribution and 'appearance were typical of Thorotrast deposi- tion. A liver-spleen scan with technetium 99m sulfur colloid showed two large defects, one in the right lobe and the second in the left lobe of the liver (Fig. 2). A radionuclide scintiangiogram demonstrated that the left-lobe lesion was hypervascular but there was no perfusion to the right-lobe lesion. The features of the left-lobe lesion were considered consistent with a hepatocellular carcinoma or a very vascular tumour such as a hemangiosarcoma, while those of the right-lobe lesion were considered consistent with a necrotic tumour or an avascular tumour such as a cholangiocarcinoma (known to be associated with Thorotrast). Cytologic study of the ascitic fluid did not reveal malignant cells. Since it was apparent that the patient was in hepatic failure and in the terminal stage of his illness, no further invasive investigation was considered. He became progressively more lethargic, then confused, and finally comatose. He died 10 days after admission. tended by 2000 ml of serosanguineous fluid. The liver was enlarged and firm, and weighed 1800 g. The capsular surface was irregular, with prominent small and large nodules. There were multiple small hemorrhagic and necrotic foci on the sur- face of the right lobe, which represented the sites of bleeding. Sections of the surface of the left lobe revealed multiple pale tan, well defined, firm nodules of hepatocellular carcinoma ranging in diameter from a few millimetres to 3 or 4 cm; between the nodules were areas of fibrosis (Fig. 3). The right lobe contained several areas of ill defined reddish, hemorrhagic and necrotic FIG. 3-Coronal section of liver and spleen. Liver is hypertrophic and firm, and contains multiple nodules of hepatocellular carcinoma and angiosarcoma (darker areas). Spleen is small and atrophic, with normal architecture replaced by fibrosis. Autopsy findings The abdomen was markedly dis- FIG. 2-Liver-spleen scan with technetium 99m sulfur colloid, showing two large defects, one in right lobe and second in left lobe of liver. 4a r FIG. 4-Hepatocellular carcinoma: clusters of neoplastic hepatocytes arranged in well defined nodules and in trabeculae (hematoxylin-eosin [H-El; X 190). 290 CAN MED ASSOC J, VOL. 128, FEBRUARY 1, 1983 lesions the largest measuring 10 cm in its greatest dimension. There was focal involvement of capsular surfaces by the vascular, friable and necrotic tumour. There was no evi- dence of portal vein thrombosis. The other abdominal organs were free of tumour. Microscopy of the sectioned liver showed a mixture of well differen- .v. V . .4 .. 4:i. . . '. >. 4 . #* . .' KA FIG. 5-Angiosarcoma: dilated sinusoids lined by increased number of hyperchromatic endothelial cells; clusters of hepatocytes and erythrocytes within lesion (H-E; >40). tiated to poorly differentiated hepatocellular carcinoma in the left lobe (Fig. 4). Bile formation was present in the tumour cells. Portal fibrosis and bile duct proliferation were prominent throughout the lobe. The right lobe showed a spectrum of morphologic changes characteristic of angiosarcoma. In some areas there were minimal changes, such as hyperplasia and hypertrophy of hepatocytes, sinusoidal dilatation and compression atrophy of adjacent parenchyma. In other areas the hepatic plates were more than one cell thick and showed proliferation of abnormal endothelial cells (Fig. 5); these cells were pleomorphic and had large irregular nuclei and prominent nucleoli, with conspicuous mi- toses. The fully developed lesions consisted of peliotic areas (bloodfilled lacunae) containing numerous sarcomatous cells and reactive hepa- tocytes. Hemorrhage and necrosis were prominent. Throughout the parenchyma of both lobes abundant Thorotrast aggregates of various sizes were identified. They were more numerous in portal tracts and areas proximal to the tumours. The spleen was markedly atrophic and weighed only 20 g. Its normal architecture was completely obliterated by fibrosis (Fig. 3). Microscopy showed numerous aggregates of Thorotrast particles in all parts of the parenchyma. A peripheral scar carcinoma 2 cm in diameter was found in the lower lobe of the left lung near the interlobular septum (Fig. 6). The pleura overlying the lesion was thickened and indented. Microscopy revealed moderate to poorly differentiated adenocarcinoma. Small aggregates of Thorotrast were present near the tumour, though they were less abundant here than in the liver and spleen; they were also noted in the hilar lymph nodes. Small cortical adenomas were identified in both kidneys. Other changes in the kidneys were compat- ible with hypertension and diabetes. Discussion FIG. 6-Small, ill defined adenocarcinoma of left lung. Scale in centimetres. Thorotrast was first used as a contrast medium for diagnostic radiology in 1928. It could be administered intravenously or intra-arterially and was used in cerebral angi- CAN MED ASSOC I, VOL. 128, FEBRUARY 1, 1983 291 ography, hepatosplenography, retrograde pyelography, hysterosalpingography and paranasal sinus cavity visualization.'8 Because of the excellent contrast it provided, it was employed in many parts of the world. Doses varied according to the diagnostic procedures. Thorotrast contains a radioactive isotope, 2.Th, which is predominantly an a-particle emitter. It has a biologic half-life of 400 years. After intravenous injection 70% is taken up by Kupffer cells in the liver. Its distribution is inhomogeneous, aggregates of various sizes being formed. It has been estimated that with a 25-ml intravascular dose of Thorotrast the liver receives a mean radiation dose of 25 rad/yr.8'9 Dahlgren'0 has suggested that three criteria be met before Thorotrast is implicated as a cause of neoplasia: first, thorium dioxide particles must be present in the immediate vicinity of the tumour; second, the latent period must have been sufficiently long (average 20 years); and, third, the dose must have been sufficiently high. The case we have presented conforms to these criteria. The long-term side effects of Thorotrast include a granulomatous reaction at the site of injection, most frequently occurring 4 to 6 years after administration. The most important late effect is the development of malignant tumours, usually in the liver. Hepatocellular carcinoma, cholangiocarcinoma and angiosarcoma have all been described.2'3"'.'6 In addition, there have been individual case reports of mesotheliomas of pleura and peritoneum, sarcoma of the soft tissues of the neck and of bone, and blood dyscrasias.5 The latent period for the development of any Thorotrastassociated tumour averages 29 years. Hepatocellular carcinoma usually produces no symptoms or signs until in its late stage, but in Western countries this stage is often preceded by cirrhosis. A rapid and unexplained deterioration in the condition of a patient with cirrhosis, along with a rapid increase in liver enzyme levels in the serum, suggests a hepatocarcinoma. Serum AFP lev- els exceeding 500 (usually 1000 to 1 million) ng/ml are common in cases of hepatocellular carcinoma.'7 The production of AFP tends to be associated with lower age, male sex, poor differentiation of the tumour and the presence of HB5Ag in the serum. 17,18 It is not clear whether the 10% to 15% of hepatocellular carcinomas that do not produce AFP are biologically different or whether some other factors are involved. Hepatic angiosarcoma is relatively rare and not Thorotrast-specific: it is also associated with long-term exposure to arsenical and copper compounds and to vinyl chloride during its polymerization to polyvinyl chloride.'7"9 However, an epidemiologic study in the United States for the years 1964 through 1974 indicated that the incidence of Thorotrast-induced hepatic angiosarcoma was increasing in the early' 1 970s and that in a larger proportion of the more recent cases there had been a relatively low dose and a prolonged latent period.20 A Danish study showed that the incidence of lung carcinoma was higher in patients given Thorotrast than would be expected in a normal population.5 On the other hand, a Japanese study found lung cancer to be more frequent in a control group than in a group given Thorotrast.2' Since the 1 950s Thorotrast has not been used and a large proportion of the population given this material has died. There remain, however, patients who by virtue of the long period since the administration of Thorotrast are at high risk of not only a single malignant disease but two or three kinds of neoplasms in the liver, as has recently been documented.4""22'23 In addition, as our case shows, neoplasms may develop at other sites too. we thank Dr. Jonathan F. Meakins for helpful discussions about his patient. References 1. DA MOnA LC, HORTA J DA S. TAVARES MH: Prospective epidemiological study of Thorosrass'exposed patients in Portugal. Environ Re's 1979; 18: 152-172 2. POPPER H, THOMAS LB. TELLES NC, FALK H, SELIKOFF IJ: Development of hepatic angiosarcoma in man induced by vinyl chloride, Thorotrast and arsenic. Comparison with cases of unknown etiology. Am J Paihol 1978; 92: 349-369 3. SMORON GL, BATTIFORA HA: Thorotrast'induced he' patoma. Cancer 1972; 30: 1252-1259 4. WEGENER K, LEIPOLz-ANGERMuLLER S: Double tumours of the liver following intravenous Thorotrast injection. Virchows Arch [Pathol Anahl 1979; 382: 6371 5. FABER M: Twenty-eight years of continuous follow-up of patients injected with Thorotrast for cerebral angiography. Environ Res 1979; 18: 37-43 6. HARR1ST TJ, SCHILLER AL, TRELSTAD RL, MANK1N HJ. MAYS CW: Thorotrast-associated sarcoma of bone: a case report and review of the literature. Cancer 1979; 44: 2049-2058 7. ABBATT JD: History of the use and toxicity of Thorotrast. Environ Res 1979; t8: 6-12 8. KAUL A, NOFFZ W: Tissue dose in Thorotrast patients. Health Phys 1978:35: 113-121 9. GRAMPA G: Radiation injury with particular reference to Thorotrast. In SOMMERS Sc (ed): Pathology Annual, vol 6. ACC, New York, 1971: 147-169 10. DAHLGREN 5: Late effects of thorium dioxide on the liver of patients in Sweden. Ann NY Acad Sci 1967; 145: 7 18-723 11. WINBERO CD, RANCHOD M: Thorotrast induced hepatic cholangiocarcinoma and angiosarcoma. Hum Pathol 1979; 10:108-112 12. RUBEL LR, ISHAK KG: Thorotrast-associated cholangiocarcinoma: an epidemiologic and clinicopathologic study. Cancer 1982; 50: 1408-1415 13. SARAGOCA A and 13 others: Thorium dioxide and the liver in 200 patients. Clinical, pathological, biochemical and immunological studies (abstr). Gastroenterology 1980; 79: 1123 14. MACMAHON HE, MURPHY AS, BATES MI: Endothelial-cell sarcoma of liver following Thorotrast injections. Am J Pathol 1947; 23: 585-612 15. BASERGA R, YOKOO H, HENEGAR GC: Thorotrastinduced cancer in man. Cancer 1960; 13: 1021-1031 16. SUCKOW EE, HENEGAR GC, BASERGA R: Tumors of the liver following administration of Thorotrast. Am J Pathol 1961; 38: 663-677 17. VOGEL CL, PRIMACK A, MCINTIRE KR, CARBONE PP. ANTHONY PP: Serum alpha-fetoprotein in 184 Ugandan patients with hepatocellular carcinoma. Clinical, laboratory, and histopathologic correlations. Cancer 1974; 33: 959-964 18. RUOSLAHTI E, SALASPURO M, PIHKO H, ANDERSSON L, SEPPALA M: Serum a-fetoprotein: diagnostic significance in liver disease. Br Med J 1974; 2: 527-529 19. TELLES NC, THOMAS LB. POPPER H, ISHAK KG, FALK H: Evolution of Thorotrast-induced hepatic angiosarcomas. Environ Res 1979; 18: 74-87 20. FALK H, TELLES NC, ISHAK KG, THOMAS LB. POPPER H: Epidemiology of Thorotrast-induced hepatic angiosarcoma in the United States. Ibid: 65-73 21. MORI T, KATO Y, SHIMAMINE T, WATANABE 5: Statistical analysis of Japanese Thorotrass-administered autopsy cases. Environ Res 1979; 18: 23 1-244 22. KOJIRo M, KAWANO V. KAWASAKI H, NAKASHIMA T, IKEZAKI H: Thorosrast-induced hepatic angiosarcoma, and combined hepatocellular and cholangiocarcinoma in a single patient. Cancer 1982; 49: 2161-2164 23. SCULLY RE, GALDABINI JJ, MCNEELY BU (eds): Case records of the Massachusetts General Hospital. Case 15-1981.NEnglJMed 1981; 304: 893-899 BOOKS This list is an acknowledgement of books received. It does not preclude review at a later date. AMERICAN PSYCHOLOGY SINCE WORLD WAR II. A Profile of the Discipline. Albert R. Gilgen. 273 PP. Greenwood Press, Westport, Connecticut, 1982. $29.95 (US). ISBN 0-313-23027-7 APPLIED THERAPEUTIC DRUG MONITORING. Volume 1: Fundamentals. Edited by Thomas P. Moyer and Roger L. Boeckx. 241 pp. lIlust. The American Association for Clinical Chemistry, Washington, DC, 1982. Price not stated. ISBN 0-91527419-1 CARE FOR THE DYING AND THE BEREAVED. Edited by Ian Gentles. 160 pp. Anglican Book Centre, Toronto, 1982. $9.50, paperbound. ISBN 0-999030-85-8 continued on page 344 292 CAN MED ASSOC J, VOL. 128, FEBRUARY 1, 1983