There appears to exist no fixed relationship between the clinical severity of yellow fever virus infection and the amount of damage found in the liver at death (9), and in some, though infrequent, instances, the liver changes in rapidly fatal cases are so few as to make diagnosis difficult or impossible. On the other hand, the content of complement-fixing antigen in the blood is roughly parallel to the severity of the infection (1–3) and hence demonstration of the antigen provides a means of diagnosis (2, 21).
The lack of correlation between the antigen content of the serum and the intensity of the lesions in the liver indicates that the antigen does not arise as a result of the visible damage produced. The similarity or practical identity of the lesions produced by the virus and by tannic acid, and the absence of the yellow fever antigen in animals succumbing to the toxic effects of the acid, support this view, and also indicate that the antigen is specific.
Nevertheless, there is some relationship, even though not absolute, between the demonstrable tissue destruction and formation of antigen. Thus, in the monkey and the marmoset, some pathologic change, however minimal, was as a rule observed in the liver, but this was not always true of the other organs. In the mouse, infection by the intraperitoneal route results in the eventual appearance of the antigen in the blood and in equally high or greater concentration in the brain, where the virus exerts its obvious damage, while the liver, which microscopically appears to be unaffected, contains no antigen, or only traces referable to the blood content of the organ (22). Just as the damage provoked in the liver of primates by tannic acid and by the yellow fever virus may be indistinguishable, there is nothing to distinguish the lesions produced by this virus in the brain of the mouse from those caused by other viruses; nevertheless, the antigen produced in the blood of the host infected with the yellow fever virus serves to differentiate this infection from similar conditions of different etiology (3).
The yellow fever complement-fixing antigen, therefore, appears to be biologically specific and the present evidence indicates that its demonstration constitutes a more accurate and reliable method of diagnosis than does histopathologic examination of tissue.
From the Serviço de Estudos e Pesquisas sôbre a Febre Amarela, Rio de Janeiro, Brazil.