In a large pertussis vaccine efficacy trial in Germany, vaccinees and/
or their family members were seen if a cough illness of >14 days was r
eported. Evidence of recent Bordetella pertussis infection included a
positive culture and/or polymerase chain reaction (PCR) and/or signifi
cant antibody values in agglutination and/or ELISA assay. From July 19
91 through February 1994, 246 adults were evaluated and 64 had evidenc
e of B. pertussis infection; of these, 38% had whooping, 26% had a his
tory of previous pertussis, and 48% were the primary cases in a family
. The 64 adult cases suggest an adult attack rate in this population o
f 133 per 100,000 population per year. Since pertussis has been endemi
c and epidemic in Germany during the last two decades, it would seem l
ikely that few persons would escape B. pertussis infections during chi
ldhood. In this regard, none of the serological controls lacked antibo
dy to all four B. pertussis antigens (lymphocytosis-promoting factor,
filamentous hemagglutinin, pertactin, and fimbriae-2). Thus, serologic
al evidence of past infection may not indicate protection, and the wid
ely held belief that individuals who have had infections with B. pertu
ssis have lifelong clinical immunity to this disease is probably wrong
.