The objective of this study was to compare 2 strategies for the management of dairy cows having retained fetal membranes (RFM) with regard to clinical traits, milk yield, and reproductive performance. In contrast to recent studies evaluating optimal strategies for the management of cows with RFM, this trial was conducted on small dairy farms with 26 to 166 cows per herd. In the systemic (SYS) group (n = 116), cows having RFM and a rectal temperature > or = 39.5 degrees C were treated with 1 mg/kg of body weight of ceftiofur on 3 to 5 consecutive days. The RFM cows without fever remained untreated. In the intrauterine (IUT) group (n = 115), all RFM cows received an intrauterine treatment with 6 g of tetracycline on 3 consecutive days combined with an attempt to remove the fetal membranes manually. The IUT cows with a fever received an additional systemic treatment with 10 mg/kg of body weight of amoxicillin on 3 to 5 consecutive days. Body temperature, daily milk yield, prevalence of vaginal discharge 28 to 35 d in milk (DIM), and reproductive performance traits within 200 DIM were monitored. The proportion of cows experiencing fever within 5 d after enrollment was greater in SYS compared with IUT. The proportion of cows with mucopurulent or purulent vaginal discharge 28 to 34 DIM did not differ between the groups. Furthermore, no significant differences between groups were found in daily milk yield in the first 10 d after enrollment, or in reproductive performance or proportion of cows culled. Significant differences in the proportion of cows with a fever in SYS and IUT have not been reported in studies with similar study designs conducted on large dairy farms. Further results on milk yield and reproductive performance, however, support findings that a management strategy for RFM based on a selective systemic treatment of feverish cows is at least as efficacious as a strategy based on intrauterine treatments of all cows and a systemic antibiotic treatment of feverish cows.