Five hundred twenty asymptomatic women attending a pregnancy counseling clinic were interviewed and screened for cervical Chlamydia trachomatis infection before therapeutic abortion. Overall, 56 (10.8%) women were culture positive for C. trachomatis. Restricting testing to women with risk factors for infection or who had cervical leukocytosis on Gram staining of cervical mucus, or who were seropositive for C. trachomatis antibodies proved impractical because of insensitivity or nonspecificity. We evaluated three chlamydia diagnostic tests to determine which test had the best performance characteristics. In comparison with culture, the direct fluorescent antigen test had a sensitivity of 89% and a positive predictive value of 78%, whereas the enzyme immunoassay had 96% and 69%, respectively. When analysis was redone with any two positive laboratory tests to define true infection status, the performance parameters of enzyme immunoassay were sensitivity 98%, specificity 98%, and positive predictive value 87%. We conclude that all women undergoing therapeutic abortion require testing for C. trachomatis infection and that enzyme immunoassay is the most effective and practical test in this group of women.
PIP: In approximately 10% of women, therapeutic abortion is the precipitating factor for acute salpingitis, and 0.5-2% of women develop acute salpingitis after the procedure. A major cause of acute salpingitis in this situation is Chlamydia trachomatis infection, with studies having found 20-34% of C. trachomatis-infected women to develop postoperative infections of the upper genital tract. The detection and treatment of chlamydial infection in such a context are therefore of significant importance. C. trachomatis infections are best detected through the isolation and identification of the organism in cell culture. Culture methodology, however, is labor intensive and the organism may not survive transportation to the laboratory because of difficulties in maintaining a suitable cold chain. Two techniques are commercially available for the nonculture detection of C. trachomatis. This paper reports findings from the clinical valuation of their performance. 520 asymptomatic women attending a pregnancy counseling clinic were interviewed and screened for cervical C. trachomatis infection before therapeutic abortion. 56 were culture positive for C. trachomatis. In comparison with culture, the direct fluorescent antigen test had a sensitivity of 89% and a positive predictive value of 78%, while the enzyme immunoassay had respective values of 96% and 69%. When analysis was reconducted with any two positive laboratory tests to define true infection status, the performance parameters of enzyme immunoassay were sensitivity 98%, specificity 98%, and positive predictive value 87%. The authors believe that all women undergoing therapeutic abortion should be tested for C. trachomatis infection and that enzyme immunoassay is the most effective and practical test for that group of women.