Impact of epidemic cholera in a previously uninfected island population: evaluation of a new seroepidemiologic method.

During an investigation of a 1982 cholera outbreak in Truk, an area without endemic cholera, 254 post-outbreak serum specimens were collected from ill and well inhabitants of a single island. These were compared with 57 specimens collected in Truk in 1964, when heat-labile toxin-producing enterotoxigenic Escherichia coli was presumably endemic but cholera did not exist. The serum was tested for vibriocidal antibody and antitoxic antibodies to cholera toxin and heat-labile toxin and the ratio of the anti-cholera antitoxin to the anti-heat-labile antitoxin was calculated. The prevalences of elevated anti-cholera antitoxin and anti-heat-labile antitoxin levels were similar in both years: an elevated anti-cholera antitoxin was found in 30% of the 1964 and 80% of the 1982 specimens, while an elevated anti-heat-labile antitoxin was found in 40% (1964) and 84% (1982). In contrast, vibriocidal antibody titers were elevated in none of the 1964 but in 64% of the 1982 specimens, and the anti-cholera/anti-heat-labile antitoxin ratio was elevated in 4% (1964) and 42% (1982). The anti-cholera/anti-heat-labile antitoxin ratio appears to differentiate the antitoxic responses to Vibrio cholerae O1 and heat-labile toxin-producing enterotoxigenic E. coli, even in areas where both are prevalent, and should serve as a useful seroepidemiologic tool.