Herpes simplex virus testing of an obstetric population with an antigen enzyme-linked immunosorbent assay

Commercial herpes simplex virus antigen enzyme-linked immunosorbent assay kits were compared to conventional culture for herpes simplex virus with 3237 genital specimens from obstetric patients. These rapid enzyme-linked immunosorbent assay tests had a sensitivity of 34.3% and specificity of 98.1% with primarily cervical specimens from asymptomatic patients. Specimens from vulvar swabs had higher positive rates than those from cervical swabs from the same patient, whether symptomatic or asymptomatic with both culture and enzyme-linked immunosorbent assay. Fifty-six women had enzyme-linked immunosorbent assay false positive tests; use of enzyme-linked immunosorbent assay testing alone could have resulted in 1.7% unnecessary cesarean deliveries. Six cases of neonatal herpes simplex virus infection were identified; two of the mothers had negative cervical cultures the week of delivery preceded by positive vulvar cultures.