Diagnosis of chlamydial infection in women attending antenatal and gynecologic clinics.

Two antigen detection systems (MicroTrak [MT], Syva Co., Palo Alto, Calif.; and Chlamydiazyme [CZ], Abbott Laboratories, North Chicago, Ill.) were compared with semiquantitative culture for diagnosis of chlamydial infection in 1,059 patients. Cultures were done on microtiter plates and blind passaged once. Culture-negative but CZ- or MT-positive specimens were recultured. True positives were positive by either initial or repeat cultures. Of 827 nonpregnant and 231 pregnant patients, 9.1 and 12.1%, respectively, had positive cultures. Overall sensitivity of the initial culture was 48.5% without passage and 86.4% with passage. The sensitivity of CZ was 67%. The sensitivity of MT in our laboratory was 50%; however, further review of these specimens by Syva employees gave a combined sensitivity of 71.6%. MT and CZ were more sensitive for pregnant patients (MT, 84.6%; CZ, 85.7%) than for nonpregnant patients (MT, 65.5%; CZ, 60.0%). All the tests had specificities above 95%. Of the specimens that were positive after initial culture without subculture, MT-negative specimens had a mean of 3.7 inclusions in culture, and MT-positive specimens had a mean of 24.8 (P = 0.002); CZ-negative specimens had a mean of 4.3 inclusions, and CZ-positive specimens had a mean of 20.0 (P = 0.026). In addition, cultures of specimens from pregnant patients had more inclusions than did those from gynecology patients, but this was not statistically significant (P = 0.096). No method is ideal; however, MT and CZ were less sensitive than was this culture system for detecting chlamydial infection in patients in gynecology clinics and were of comparable sensitivity for pregnant patients.