Yellow Fever Accidentally Contracted in the Laboratory

Summary and ConclusionsThe results of studies made in seven cases of yellow fever acquired in the laboratory have been presented. The group embraced infections ranging from very mild to moderately severe. All patients recovered. Details of the clinical courses are provided because the infections furnish examples of typical mild yellow fever, the form of the disease rarely recognized yet paramount in importance from the standpoint of epidemiology and public health. In the mode of onset the light attacks corresponded with classical descriptions. Yet of the signs of the disease commonly recognized as cardinal, i.e., paradoxical pulse-temperature relationship, jaundice, albuminuria, and "black vomit," only the first was regularly observed, while the others were absent or present in minimal form.Bradycardia was shown to occur independently of jaundice. Additional evidence of myocardial injury was brought out by bedside observations, by x-ray measurements of acute dilatation of the heart, and by electrocardiograms. By the last method of examination, furthermore, abnormalities similar to those reported in the experimental disease in monkeys were shown to occur in our patients.The demonstration of guanidine increase in the blood of one of the mildly infected patients links yellow fever to other diseases also characterized by extensive liver destruction.Convalescent human serum, in amounts of 5 cc. at bimonthly intervals, was ineffective in preventing laboratory infection.A systematic study of the leucocytic reaction was conducted in five of our cases. Following the onset of disease, a progressive leucopenia, due chiefly to a decrease in neutrophiles, developed. The cell count attained its lowest point on the fifth or sixth day. At this time, the lymphopenia present at the initiation of symptoms was already disappearing. The monocyte count, unaltered during the period of acute illness, rose during convalescence. The most striking leucopenia occurred in two of the milder cases. The reaction of the leucocytes in each case showed sufficient constancy to suggest that it may be an important diagnostic aid.The diagnosis was established by the successful transfer of the virus to monkeys and to mice and by the demonstration of protecting antibodies in each patient's blood. Virus was shown to exist in the blood of one patient as late as the fifth day (107 hours after the onset) of disease, and neutralizing antibodies were demonstrated in the same case as early as the fourth day (83 hours). Thus, the simultaneous presence of virus and antibody in human blood in yellow fever has been established.