Vaginal colonization with Escherichia coli is an integral step in the development of acute cystltis, and persistent vaginal coliform colonization may also be a predisposing step to recurrent urinary tract infections. For this reason, we evaluated antibiotic concentrations in the vaginal fluid, serum, and urine and the vaginal colonization by E. coil of 56 women receiving either ofloxacin (200 mg orafly twice a day) or trimethoprimsudfamethoxazole (TMP-SMX) (160/800 mg oraHly twice a day) for the treatment of acute cystitis. Ofloxacin and trimethoprim both penetrated into vaginal fluid to a considerably greater extent than sulfamethoxazole. Among 33 patients given ofloxacin, the concentration of the drug in vaginal fluid during one dosage interval ranged from 1.6 to 21.6 >g/ml. In 21 women given TMP-SMX the range of drug concentrations in vaginal fluid was 2.6 to 32.5 ,g/ml for TMP and 1.0 to 6.2 ,ug/ml for SMX. Treatment with both ofloxacin and TMP-SMX remarkably reduced vaginal colonization by E. coil during and up to 30 days after therapy. For the ofloxacin-treated women, eradication of vaginal E. col was aated with a hig ratio of drug concentration in vaginal fluid to that in serum. We conclude that ofloxacin and TMP both achieve hig concentrations in vaginal fluid and are equaHly successful in eradicating E. coil from the vagina.Vaginal colonization with Escherichia coli is an integral first step in the pathogenesis of acute cystitis, and persistent vaginal colonization with E. coli has been associated with the development of recurrent urinary infections in women (2,7,13 Vaginal fluid was obtained by using a sterile preweighed 25-mm-diameter filter paper disk (no. 42; Whatman, Inc., Clifton, N.J.) that was stored in a sealed test tube. During the speculum examination, the disk was placed high on the vaginal wall for 15 s, with care taken to avoid introital contamination. The disk was removed with sterile forceps and placed into the original test tube, and the tube was sealed tightly and stored at -70°C until analysis. A vaginal culture was also obtained by using a sterile cotton-tipped swab.Andbiotic therapy. As previously described (3), patients were randomly assigned to receive either oral ofloxacin (200 * Corresponding author. mg) or TMP-SMX (160/800 mg) twice a day for 3 to 7 days. Patients were given the first dose in the clinic and were instructed to take their assigned doses approximately every 12 h. Diaries were used by the patients to record the time of each dose.Foflow-up visits. Patients returned to the clinic during therapy at day 2 or 3 and at days 5 to 9 and 30 posttherapy. During these visits, vaginal and urine cultures were again obtained. Blood and vaginal fluid samples for determination of antibiotic concentrations were collected at the first follow-up visit. All specimens were collected within 15 min of each other for each patient.Bacteriology. All cultures and susceptibility tests were performed in our laboratory as previously described (8 (82:18, vol/vol). No interferences from either e...