From December 1947 through March 1948 a severe epidemic of encephalitis, affecting natives, Americans and other nonresidents, occurred on Guam. The epidemic was found to be due in part to JBE virus and in part to mumps virus. This was the first recognized outbreak of JBE on Guam or in any other truly tropical area. Mumps was occurring in an unusual epidemic form for it had been absent from the island for 6 or 8 years, and from one part for about 17 years. A total of 1,647 cases were reported in natives, a rate of 66.6 per 1,000. The total encephalitis rates were 1.86 per 1,000 among natives and 0.15 among the military, with somewhat similar low rates among various groups of nonresident civilian personnel. JBE virus was isolated from four of five fatal cases.
By using serological tests in addition to virus isolations it was determined that, of those cases which could be specifically diagnosed, approximately one-third had been infected by JBE virus only, one-third by mumps virus only, and one-third more or less simultaneously with both viruses. Only six cases could not be classified. The time of the peak of the mumps epidemic and of that of JBE coincided almost exactly. All fatal cases had JBE virus infection, one had mumps also. Serological patterns were observed to vary considerably between individuals but certain common patterns were noted.
The native cases with JBE virus infection had their highest incidence under 5 years of age. Males were more frequently affected than females. Cases occurred predominately in the southern portion of the island where there were few Americans and other nonresidents.
Climatologic data were intensively studied. Attention was focused on possible associations between climate and the possible dissemination of virus by a mosquito vector. Several climatologic aspects were unusual but a marked prolongation of the normal rainy season may have contributed to the epidemic. Findings from mosquito surveys, together with other data, pointed to C. annulirostris as the probable vector. The climatologic conclusions are compatible with its known breeding habits.
Serological surveys were made at the end of the epidemic on a number of population groups and on wild and domestic vertebrates. Samples of native children were bled again a year later (1948–49) and again in 1953 and in 1957. Interpretation of the serology is difficult and has been influenced by certain events and observations summarized in part as follows. Dengue was endemic until the vector A. aegypti was virtually eradicated in 1945 and 1946; then dengue disappeared as a clinical disease. No further proven cases of JBE encephalitis and no epidemic of a similar disease occurred subsequent to the epidemic studied. No dengue or JBE antibodies were found in any children born subsequent to the epidemic, yet no mosquito population or species decrease has been recognized since the eradication of A. aegypti, which was prior to the encephalitis epidemic.
Serological findings at the end of the epidemic included the following salient features. Varying proportions of wild and domestic mammals and birds, as well as man, had developed neutralizing antibodies. In man there were more persons with neutralizing antibodies than with CF antibodies except below 5 years of age. This excess of those with neutralizing antibodies increased with age up to adulthood. This same difference between the proportion with CF and neutralizing antibodies was observed in areas of low and high current infection rates. Total CF antibody prevalence rates in southern villages were essentially 50 per cent just at the end of the epidemic. Two to 3 months later they had fallen to 14 per cent. Neutralizing-antibody titers were highest to JBE virus, lower to WN and lowest or absent to SLE virus. Fewer were tested for antibodies to MVE and these were negative or much lower in titer than to JBE. Sera collected in 1953 were tested with several other group B viruses. The only positives were in persons born before 1947 and these persons also had antibody to dengue viruses. CF antibodies to other group B viruses were frequently present. When included, SLE antigen gave elevated serum titers more frequently than the other antigens, but usually lower than those with JBE. It was concluded that JBE and dengue viruses were probably the only group B agents necessary to consider and that no one had been infected with dengue for about 2 years before the encephalitis outbreak.
JBE neutralizing antibodies without CF were considered to be due to one of three possibilities: (1) recent JBE infection, following which CF antibody had persisted only 1 to 3 months and had just been lost; (2) repeated previous dengue infections; or (3) infection with JBE virus which had occurred during previous years. Either of the first two interpretations supported the hypothesis that JBE virus had just invaded the island, and the last, that the virus had been established there for at least a few years. All evidence has been weighed in support of these two hypotheses and it is suggested that JBE virus probably invaded the island de novo as did mumps, and that this was a “virgin” outbreak. The virus then disappeared, apparently due to the absence of some factor necessary to permit its maintenance at an endemic level. The age distribution of cases may be that expected in a virgin outbreak but, since this is the first virgin epidemic on record, the age distribution is accepted with caution. Dengue possibly afforded some clinical protection and may have influenced the age pattern. Infection rates with JBE virus in certain villages probably attained a level of at least 50 per cent or possibly even 80 per cent, figures never previously approached in other reported outbreaks. It was observed that although A. albopictus was introduced into the island and has replaced A. aegypti, dengue nevertheless did not persist or recur.
Department of Epidemiology and Microbiology, The Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania.
The Walter Reed Army Medical Center, Washington, D. C.
On leave of absence from the Division of Medicine of the University of California and the Viral and Rickettsial Laboratory of the California State Department of Public Health.