Two groups in Denmark and Sweden, totalling 943 healthy women in the first trimester of pregnancy who were undergoing elective abortion, were examined in a prospective study. Group 1 consisted of 432 women from Denmark, all examined prior to and 2 weeks after the operation. At the first examination, samples were taken from the cervix for the isolation of chlamydiae and gonococci. Chlamydia trachomatis was isolated from 23 (5.3%) of the women. Postoperatively, 15 patients developed acute pelvic inflammatory disease (PID). Five of these patients harbored C trachomatis before the operation, and all 5 developed a significant change in the titer of immunoglobulin G microimmunofluorescence antibodies to the organism. In group 2, 25 (4.9%) of the 511 women from Sweden harbored C trachomatis before the intervention. Patients with chlamydiae were treated, before or in conjunction with the operation, with antibiotics active against chlamydiae. None of the women with cultures positive for chlamydiae showed signs of PID postoperatively. The difference in the postoperative frequency of PID in chlamydia-positive patients between groups 1 and 2 was significant (2P = .0393, Fisher exact test). Gonorrhea was diagnosed in 1% of the women in both groups. Only 1 of these patients developed PID postoperatively, but no pili antibodies to Neisseria gonorrhoeae were found during the course of the disease. It is concluded that women applying for abortion should be examined and treated not only for gonorrhea but also for infection with C trachomatis either before or, at the latest, in conjunction with the abortion.