9 Oktober 1965 S.A. TVDSKRIF VIR GENEESKU'DE 817 the academic field to ensure that the future generations of be made for the trivialities which are so important to the medical practitioners a~e imbued with the spirit of ser­ individual. The pain that starts in the pit of the stomach, vice; and it is the responsibility of the administrators, medi­ goes between the shoulders and up to the top of the head cal and lay, to ensure that the systems evolved enable medi­ until 'it is pressing like a ton weight doctor!' is very real cal practitioners to carry out the service they are anxious and important to the patient. In the vast organizational and willing to render. Like Peter Medawar I do not think aspects of modern medical services the patient, the reason one can in all honesty prescribe for the future. Historic­ for such organization, is apt to be overlooked instead of ally it was only in the middle of the last century that a being the central feature and chief beneficiary of all the physical examination of the patient was accepted as an manifold skills and wonders of the new medicine. The aid to diagnosis. Before that only the symptoms as given touch of human kindness, reassurance and comfort are the by the patient and the opinion of the patient himself were tried and true remedies of the old medicine, it is for us to taken into consideration. Over the first half of this century ensure they have a large place in the new medicine. the physical examination has been given great prominence, Let me end with the words of John Ellis, spoken in the it is expected by the patient. No private or hospital patient Hasting Lecture last September: 'The new medicine must would consider himself fairly treated unless he was bring comfort and relief and give support when nothing examined by the doctor. The first half of this century, too, more can be done and still stand guard over the more has seen developed the laboratory examination as part of trivial complaints'. the diagnostic procedures, and now the clinician is calling more and more for assistance from the scientist and is REFERENCES himself undertaking more biochemical and physiological I. Pickering, G. W. (1963): Brit. Med. J., I, 133. 2. McFarlane, B. (1964): Ibid., 2. 1091. procedures. The chemist, the physicist, and the mathemati­ 3. Ellis, J. R. in Reid, J. V. O. and Wilmot, A. J. eds. (1965): Medical Education in South Africa~ p. 88. Pietermaritzburg: Natal University cian, including the statistician, are today in the forefront Press. of medical research and advanced medical thinking. 4. Kench, J. E. in Reid. J. V. O. and Wilmot, A. J. eds. (1965): Ibid., p. 44. Now, as we enter the second half of the century, will 5. WHO E<pert Committee on Genetics (1962): Wld. Hlth. Org. Techn. Rep. Ser., no. 238. I. the machine take over teaching? Will the computor make 6. Kipps, A. (1964): J. S. Afr. Vet. Med. Assoc.. 4, 35. It 7. Wiener, N. (1961): Cybernetics, 2nd ed. Cambridge. Mass.: MIT. one process of investigation, diagnosis and treatment? 8. Owen, S. G., Hall. R. and Wailer, t. B. (1964): Postgrad. Med. J .. is idle to speculate; let us rather have the slogan: 'Today's 40, 59. 9. Annotation (1965): Lancet, I, 1055. skills for today's ills'. Nevertheless, provision still has to 10. Medawar. P. (1965): Proc. Mayo Clin., 40. 23. THE AETIOLOGY OF PRIMARY CARCINOMA OF THE LIVER IN AFRICA: A CRITICAL APPRAISAL OF PREVIOUS IDEAS WITH AN OUTLINE OF THE MYCOTOXIN HYPOTHESIS* A. G. OElTLE, Cancer Research Unit of the National Cancer Association of SOUTh Africa, SOUTh African InstilUTe for Medical Research, Johannesburg 'Medical research in the tropics is still capable of Although the complete web is infinitely complex, in yielding harvests and there are many opportunities medicine we are concerned with effective action, and this in geographical pathology. Harvesters, not gleaners only. are still needed.' requires no more than the identification of the relevant V. Ramalingaswami' proximate influence which, if removed, would lead to a substantial reduction in the risk of the disease. On this Since the turn of the century, primary carcinoma of the empirical basis, what is diagnosed as a causal relation liver has been recognized as unusually common in the may turn out to be predisposing, carcinogenic (initiating) Cape Coloured and Bantu of South Africa.' Further or co-carcinogenic (promoting). The factors implicated studies have shown it to be common in many parts of might be several degrees away from those immediately Africa south of the Sahara, and a high incidence has responsible in carcinogenesis, but their discovery should become generally accepted as characteristic of the indige­ disclose the chain of causality whose precise links must nous races of the subcontinent. subsequently be ascertained. It seems logical to look for a single major carcinogenic In the past a number of hypotheses have been put for­ stimulus to explain this high frequency of liver cancer ward whose variety and flimsiness reflect the smattering in Africa. This does not deny the complex aetiology of of epidemiological evidence on which they were based. cancer,' nor does it exclude the possibility of other causes Correlation is not necessarily causation, and hypotheses. such as chronic alcoholism or virus hepatitis in some cases. however plausible, should be subjected to crucial tests. Nevertheless, it is better to risk the Scylla of over-simpli­ The recent discovery that potent hepatotoxins are pro­ fication than the Charybdis of over-complication, if the duced by fungi on spoiled foods has provided yet another latter becomes no more than a bewildered acknowledge­ plausible explanation of the occurrence and distribution ment of complexity that explains away anomalies, and of liver cancer in Africa and similar areas.' The evidence effectively hinders logical analysis. for this is more convincing and it fits the facts more adequately than its predecessors, but nevertheless it re­ • P,a~r present~~ at the Symposium on Mycotoxicosis organized by the mains no more than a working hypothesis, and much more Natlo~al Nutnhon Research Institute and the Department of Agricultural Tecnmcal Services, held in Pretoria on 25 February t965. critical examination and testing is required. 818 S.A. MEDICAL JOURNAL 9 October 1965 CRITICAL EXAMINATION OF THE PROBLEM 3 separate medical examinations a few months previously. 'A theory is not a tbeory unless it can be disproved Tn some the first evidence of an abnormality is the terminal by some possible experimental outcome.' intraperitoneal haemorrhage. Others, on the other hand, ... 1. R. Platt' run a chronic course similar to that familiar in the West, and all intermediate gradations are known. A. Pathology of Liver Cancer in Africa Its occurrence in young persons, the rapid clinical course, A high incidence is found to hold for one histologic as well as the annual and seasonal fluctuations, all suggest type of liver tumour only, viz. hepatocellular hepatoma a remarkable rapidity of carcinogenesis-not so rapid, including those of adenoid pattern. Carcinomas of the however, as to obliterate the differences between Bantu gallbladder or of the intrahepatic or extrahepatic bile dUc~s miners of different origin, for these persist while they are are no more common in Africa than In the West. This in employment on contracts of approximately 9 months suggests either that the agent acts specifically on the liver on the mines on identical diets and conditions of lIfe. cell, or that it is excreted in the bile in relatively inert Nevertheless, in primary carcinoma of the liver, carcin~­ form, or both. genesis seems to be shorter than in other types. of carcI­ The concept of a hepatotoxic agent is supported by the noma for it has been noted that Japanese nnrrugrants to common association of hepatoma with cirrhosis, although Haw;ii do not carry with them any of the high risk of cirrhosis of the liver is not a necessary precursor of liver liver cancer which the Japanese population in the home­ cancer. Where the incidence of liver cancer is high, an land displays." increased incidence of malignant change is found in non­ cirrhotic livers as well. The type of cirrhosis is important. The association of hepatocellular carcinoma in Africa sugg~sts In cases with post-necrotic cirrhosis seen at postmorte~ with the hyperplasia of post-necrotic cirrhosis, in Johannesburg, malignant degeneration had occurred In that the cancers follow exposure to an agent prodUCIng over 50J!0, whereas it was much less common in septal necrosis followed by reactive hyperplasia. This is analo­ cirrhosis: gous with the phenomenon of promotion in epidefI?al The proportion of cirrhotics undergoing malignant carcinogenesis. Maltoni and Prodi'2 have shown t~at clrr­ change is also higher in areas of high incidence. Thus, In hogenic treatment of rats with carbon tetrachlonde and butter yellow simultaneously, produced an earlIer appear­ Lourenco Marques the figure is 61-5'70 in males and 51-4% in females,' whereas in Johannesburg the figures are 47 ance of hepatic neoplasms than in controls treated WIth and 12% respectively.s butter yellow alone. Furthermore, after a subliminal dose of butter yellow, subsequent treatment with carbon tetra­ Both in Lourenco Marques and in Johannesburg the chloride induced the appearance of tumours. In animals proportion of elderly cirrhotics undergoing malignant de­ previously treated with carbon tetrachloride and subse­ generation is less than that in the younger group. It IS quently given butter yellow, the onset was delayed by possible that this is an artifact; e.g. it may be that elderly patients with a relatively acute disease such as Ilver cancer comparison with those treated with butter yellow al<:llle.