This field study focuses on the possible effects of a combination of nonsteroidal antiinflammatory treatment (carprofen) and a local and parenteral antibiotic on cure rates, survival rate and return to milk production of severe clinical mastitis cases. 19 cows in 3 herds (blocked by parity) with severe clinical mastitis during the first 120 d of lactation (median = 28 d) were treated with antibiotics and onehalf of these cows were treated with 1.4 mg/kg bodyweight carprofen (Rimadyl Rind®, Pfizer GmbH Tiergesundheit, Germany). Double milk samples for bacteriology were collected from clinically affected udder quarters before treatment and at 14 (± 3) and 21 (± 3) days after commencement of treatment for cytomicrobiological analysis. Additionally, every single animal was clinically examined on d 1 to 1. Treatments were initiated before culture results. There were no differences between the treatment groups in calving date, days in milk, age, breed, rectal temperature, clinical, bacteriological, cytobacteriological cure rate and in the number of cows that were defined as treatment failures (i. e. died, re-treated, relapse). Six (22.2%) vs. seven (19.4%) cows in the carprofen and control groups failed, respectively. The milk yield was significantly higher in the carprofen-treated group compared with the control group after treatment. The present work gives first indications that treatment of cows with severe clinical mastitis with a combination of carprofen and antibiotics could result in a faster return to milk production compared to treatment with antibiotics alone. If this effect can be affiliated to the administration of carprofen alone has to be examined in further studies.