Yellow Fever Endemicity in West Africa, with Special Reference to Protection Tests

Summary and Discussion

The results of all tests described above are summarized in table 5. It will be noted that among 125 sera, collected at random from persons living in Ibadan and Ilorin, in the region considered endemic, thirty-eight, or 30.4 per cent, protected duplicate animals. In Ife, in the same area, but where an epidemic of yellow fever had recently occurred, the percentage of protection was even higher, as seventeen among twenty-five specimens tested, or 68 per cent, protected two animals. The results of similar tests carried out with sera from persons resident in Northern Nigeria stand out in sharp contrast, for among ninety specimens tested only one was found to be positive.

Twenty-four per cent of the sera of children between the ages of 4 and 5 years, taken at random in Ibadan, and a similar percentage of sera taken from children between the ages of 4 and 7 years at llorin protected duplicate animals against yellow fever virus. It appears, therefore, that at least one of every four young children in these cities has had an attack of yellow fever. Although, as will be explained below, the actual number of children who have had yellow fever is probably greater than these figures suggest, the results afford strong evidence of the existence of endemicity in Southwestern Nigeria.

It should be noted here that an effort is being made to correlate the evidence for endemicity, afforded through these tests, with the identification of actual cases of yellow fever among the native population. Children in government and mission schools have been kept under fairly close observation, febrile conditions among children and adults have been carefully investigated, and large numbers of natives have been questioned as to past illnesses. As mentioned above, an extensive epidemic was observed in Ife and strains of virus were isolated from two patients. The presence of infection was also demonstrated by the mosquito catch in that town. In Ibadan and Oshogbo in 1925 and 1926 several cases were observed among Europeans and one case was seen in a native. On the other hand, it is not yet possible to identify clinically the mild cases which are presumably occurring constantly and in considerable numbers among the natives. This study is complicated by the fact that the African, being suspicious and preferring native medicine to that of the white physician, has a tendency to hide his sick. Furthermore, almost 100 per cent of children harbor malaria parasites and over 50 per cent of those in Ibadan are infected with schistosomiasis. Under the circumstances, it is difficult to eliminate malaria in febrile cases, and albuminuria loses its significance when the patient is the victim of schistosomiasis.

Histories of earlier attacks of jaundice with suggestive symptomatology were obtained from a number of persons, but the percentage of these whose sera afforded protection was not materially higher than that found in persons of the same age groups taken at random. As recently reported, (4), a disease associated with jaundice and superficially resembling yellow fever is widespread in West Africa and adds greatly to the difficulty of diagnosis in individual cases and evaluation of histories of past illnesses.

During the extensive epidemics studied on the Gold Coast, a few babes-in-arms with scleral icterus were observed, but no child under 5 years of age presented a clinical picture justifying the diagnosis of yellow fever. It is accordingly believed that the manifestations of this disease in native children, especially in endemic areas, are rarely sufficiently marked to permit of clinical diagnosis even under the most favorable conditions.

As previously mentioned, duplicate monkeys were used in testing each specimen, and in the interpretation of the results sera were considered as positive only when both animals survived and negative when either or both died. In analyzing the findings, however, it will be noted that the percentage of cases in which only one animal died while the other survived is very much higher with specimens from Ibadan and Ilorin than with those secured from cities and towns in the north: 27 in 125 as compared with 6 in 90.

The fact that rhesus monkeys are not equally susceptible and that a certain number are entirely refractory even to the highly virulent strain of virus used in these tests affords an explanation for the survival of some animals even when the sera contained no protective bodies, and where the duplicate animals died. However, as all sera in both series mentioned were tested against approximately the same amount of virus, and under similar conditions, we fell certain that the large number of sera from the south coming within this category cannot be explained on the basis of relative resistance of the Macacus rhesus alone.

It is impossible to estimate with absolute accuracy the percentage of monkeys which is totally refractory or only moderately susceptible to a particular sample of yellow fever virus. But if, for purposes of comparison, the tests with the sera from Bauchi, Zaria, and Kano are taken as controls for the experiments carried out with specimens from other localities, and we accept one as protecting and the remainder as definitely negative, it appears that of the total of 178 monkeys with which the remaining eighty-nine sera were tested, only six, or slightly over 3 per cent, survived, and of these, five had definite febrile attacks from which they recovered.

It is probable that some variation in susceptibility occurs with different lots of animals and at different times; but, making allowance for such variations, we fell confident that not more than 6 per cent of normal rhesus monkeys survive the injection of 0.1 cc. of virulent blood taken at the onset of fever from an animal infected with the Asibi strain, and this is in accord with the results obtained in a large number of uncomplicated experiments in which animals were injected with this dose of virus.

Among a total of 125 sera from persons in Ibadan and Ilorin, 38 specimens protected both animals. The remaining 87 sera were tested with 174 monkeys, and of these, 27, or 15 per cent, survived. If 10, or roughly 6 per cent, are admitted to have been refractory, there still remains a balance of 17 animals representing an equal number of sera to be considered and, though it could not be demonstrated experimentally, it is believed that some of these sera contained protective elements which, for reasons not clearly understood, afforded a partial or complete protection in one animal and failed completely in the other. Definite evidence that this does occur is shown by the fact that the serum of J., an African who had a moderately severe attack of yellow fever in 1928 and from whom a strain of virus was secured, has repeatedly protected monkeys in 1 and 2 cc. amounts against large doses of virus. On one occasion, however, when this serum was tested in 2 cc. amounts in duplicate monkeys against a uniform dose of virus, one of the test animals was fully protected, while the other died of yellow fever. It seems likely, therefore, that the percentage of persons in the Ibadan-Ilorin area who have had an attack of yellow fever is higher than is indicated in the figures.

It should also be remembered that the amount of virus contained in the routine test dose of 0.1 cc. is very much greater than we originally supposed. Titrations have recently been carried out with the blood of numerous monkeys infected with three different strains of yellow fever virus and using a 10 to 25 per cent solution of normal rhesus serum as a diluent. In these experiments it was found that amounts of between 0.0000001 and 0.000000001 cc. of blood have frequently proved fatal. The amount of blood used in these protection tests, therefore, may safely be considered as representing approximately 1,000,000 lethal doses.

The amount of immune serum used in a protection test with yellow fever virus appears to be an important factor, as the potency of the serum can be measured quantitatively. In several experiments serum from persons who had recovered from yellow fever was tested in decreasing amounts against a uniform dose of virus. The results were always clear-cut and indicated that while large doses showed full protection small amounts gave negative results. Further experiments are being made to determine the optimum dosage of serum and virus in protection tests, and it is believed that when the test has been refined we may be able to detect the presence of protective bodies in the sera of a greater number of persons who have had yellow fever than is possible with present methods.

The results of tests carried out in Kano and Sierra Leone indicate that a certain number of cases of yellow fever have occurred in these cities during recent years. As the former is an important commercial town and the terminus of the railroad passing through the endemic area, and as Freetown is in close communication with places where large epidemics have occurred within the last few years, conditions have obviously been favorable for the frequent introduction of infection, and some secondary cases have apparently developed. The contrast, however, between the recent and past history of the disease in Freetown affords a striking demonstration of the value of piped water-supply and other sanitary measures in the control of yellow fever, and is a tribute to the energy and efficiency of the British West African Medical Service.