ABSTRACT
The first well-documented outbreak of viral hepatitis E in Africa was described in 1986 in Côte d'lvoire. Subsequently, no other outbreaks have been observed in the country. Côte d'lvoire therefore offers an excellent opportunity to evaluate the prevalence of sporadic viral hepatitis E in a country where the frequency of non-A, non-B, non-C viral hepatitis appears to be high. The study was carried out in Abidjan, the most populous city, and involved 111 hospitalized patients suffering from non-A, non-B and presumed non-C acute viral hepatitis. Screening for leptospirosis or a toxic etiology was carried out and the risk of including such patients eliminated. Diagnosis of viral hepatitis A was excluded from the absence of IgM anti-HAV antibodies. Patients with HBsAg and anti-HCV antibodies were not included in the study, although co-infection in asymptomatic HBV carriers or subsequent infection in patients who had recovered from a past HCV infection remained possible. There was a risk that some patients with late appearance of anti-HCV antibodies were included since PCR tests could not be performed. Cytomegalovirus or Epstein-Barr virus was not involved, since no specific IgMs against these viruses were detectable. Large discrepancies between the two commercial enzyme-linked immunosorbent assays (ELISAs) available for serological diagnosis of hepatitis E (Abbott and Genelabs) were observed. Among the 53 sera screened using both tests, only 20 gave positive results in both, and all such sera were confirmed using a domestic immunological test involving inhibition of labelled, well-documented anti-HEV-specific human IgG. Immunological confirmation was obtained for only half of the sera with discordant results in the commercial ELISAs. Full agreement between both commercial tests was observed for only 59% of the sera studied. The minimal incidence of sporadic viral hepatitis E among hospitalized patients in Abidjan with an acute hepatitis was estimated to be 27%.
