COMMUNITY-BASED PREVALENCE PROFILE OF ARBOVIRAL, RICKETTSIAL, AND HANTAAN-LIKE VIRAL ANTIBODY IN THE NILE RIVER DELTA OF EGYPT

To determine the current prevalence of antibody to arboviruses, rickettsiae, and hantaan-like viruses, a survey was carried out in the Nile River Valley of Egypt, one of the principal foci of the 1977-1978 Rift Valley fever (RVF) outbreak. Blood specimens were obtained from 915 persons representing 190 study households. Enzyme immunoassay testing showed that the overall prevalence of IgG antibody was 4% to sand fly fever Sicilian (SFS), 2% to sandfly fever Naples (SFN), 15% to RVF, 20% to West Nile, and 4% to Hantaan (HTN) viruses. Antibody was demonstrated among 32% of the same study subjects to Coxiella burnetii, 58% to Rickettsia typhi, and 32% to R. conorii. The prevalence of agent-specific antibody tended to increase with age. Particularly notable was the low prevalence of RVF infection in children born after the height of the RVF outbreak. No detectable antibodies were found in the population less than seven years of age and in only 3% of those 7-12 years old. In contrast, 26% of the study population 13-19 years old, who were young children and infants at the time of the outbreak, were found to have RVF antibodies, suggesting that the level of intensity associated with transmission decreased considerably following the documented 1977-1978 outbreak. Geometric mean titers (GMT) ranged from 139 for C burnetii to 1,305 for RVF, and did not vary significantly by age, except for high titers for RVF in the 20-49-year-old age group. A significant upward trend in GMT was also noted when antibody was detected in the specimen for more than one phlebovirus. This was observed for SFS, SFN, and RVF, i.e., 1,172 for RVF only, 1,334 for RVF and SFN, 1,828 for RVF with SFS, and 2,1 11 for RVF with SFN and SFS (P < 0.05). These figures attest to the boosting phenomenon found when antibody for one phlebovirus is found in association with others. These findings show prevalences of rickettsial antibodies and warrant further study of disease incidence associated with acquired infection. Additionally, familial clustering of infected cases in households was evident only for C burnetii, R. typhi, and R. conorii, and was independent of sex.