Fine-tuning the fluorescent antibody test for chlamydial infections in pregnancy.

Endocervical specimens from 168 asymptomatic women in the second trimester of pregnancy were screened for Chlamydia trachomatis infection with both standard tissue culture methods and a direct fluorescein-conjugated monoclonal antibody assay (MicroTrak). Based on tissue culture results, the prevalence of Chlamydia trachomatis was 26.2% in this population. Compared with tissue culture, the monoclonal antibody assay's sensitivity, specificity, and positive predictive value varied depending upon how many elementary bodies were used to define a positive test. Specifically, if a cutoff of ten elementary bodies was used (as per the manufacturer's instructions), the sensitivity, specificity, and positive predictive value of this test were 86.3, 98.4, and 95.0%, respectively. At the other extreme, a cutoff of one elementary body produced a more sensitive but less specific test, with parameters of 93.2, 89.5, and 75.9%, respectively. Based on these data and operating characteristic analysis, the cutoff value defining a positive test was appropriately set at two or more elementary bodies, at least for this study population. This resulted in a sensitivity, specificity, and positive predictive value of 93.2, 95.2, and 87.2%, respectively. This monoclonal antibody assay appears to be a reasonable substitute for cell culture for Chlamydia trachomatis. However, because of varying magnitudes and implications of false-positive and false-negative tests, clinicians are urged to determine the appropriate breakpoint for their individual laboratories and patient populations before substituting the direct fluorescent antibody test for tissue culture.