In this field trial, a protocol for the treatment of retained fetal membranes (RFM) without any intrauterine therapy was compared with 3 protocols based on the intrauterine use of antibiotic pills (AP), the manual removal (MR) of the fetal membranes, or the combination of both (PR). The study was conducted on 5 commercial dairy farms in Germany. Cows with RFM for at least 24 h after calving were assigned to 1 of 4 treatment groups. Cows of all groups with a rectal temperature >or= 39.5 degrees C received a systemic antibiotic treatment with ceftiofur (1 mg/kg per d) for 3 to 5 consecutive days. In case of continued fever after 5 treatments, cows received a different antibiotic as an escape therapy. In the reference group (REF; n = 131), cows did not receive any additional treatment. All cows in group AP (n = 119) received intrauterine treatment with antibiotic pills consisting of 1,000 mg of ampicillin and 1,000 mg of cloxacillin for 3 consecutive days. In group MR (n = 121), an attempt was made to remove the fetal membranes manually, but uterine pills were not administered. In group PR (n = 130), an attempt was made to remove the fetal membranes manually and all cows received a local antibiotic treatment as in group AP. All cows received 2 doses of 25 mg of PGF(2alpha): one dose between 18 and 24 d and another between 32 and 38 d postpartum. Statistical analyses were performed using binary logistic regression models and survival analyses with group REF as reference. Of all animals, 79.8% had a body temperature of >or= 39.5 degrees C at least once within 10 d postpartum and were treated with ceftiofur. Occurrence of fever within 10 d postpartum was significantly lower in groups AP and PR compared with reference group REF, but was not different between groups MR and REF. Risk of receiving an escape therapy in case of fever after 5 treatments with ceftiofur did not differ among groups. Reproductive performance measures did not differ significantly between group REF and any of the comparison groups. Compared with a treatment protocol based only on systemic treatment with antibiotics for cows with a fever, neither intrauterine antibiotics nor manual removal of fetal membranes alone or in combination reduced proportions of cows needing an escape therapy nor did those treatments improve reproductive measures in the current lactation. Systemic treatment alone based on elevated rectal temperature was effective and reduced use of antibiotics compared with therapies that included intrauterine antibiotics.