Antigen detection for the diagnosis of gonorrhea.

We compared the diagnostic value of an enzyme immunoassay method for detection of gonococcal antigen in genital secretions with culture results and direct Gram stain for Neisseria gonorrhoeae in 1,171 men and 723 women attending a sexually transmitted disease clinic. When compared with culture results in men, the immunoassay provided a sensitivity of 94% and a specificity of 98% and was essentially equivalent to the urethral Gram stain. The predictive value of a positive immunoassay was 97% in men with a urethral discharge in whom the prevalence of gonorrhea was 36%, and 30% in men without urethral discharge, who had a 2% prevalence of gonorrhea (P less than 0.001). The sensitivity of the immunoassay was 95% in men with and 67% in men without urethral discharge (P less than 0.01). In women, the immunoassay resulted in a sensitivity of 78% and a specificity of 98% compared with cervical culture and had a significantly better sensitivity than the cervical Gram stain (78 versus 48%, P less than 0.001). Analysis of patients with discrepant culture and immunoassay results suggested that most culture-negative, immunoassay-positive patients probably did not have gonorrhea. After treatment, all but 1 of 59 originally culture- and immunoassay-positive patients became negative in both tests by 3 days. Results of the immunoassay were not affected by transport or by refrigeration for up to 30 days.