Effect of dry-cow therapy on subclinical mastitis--an evaluation of long-acting and short-acting intramammaria.

This field study was conducted to evaluate the effect of selective dry-cow therapy with long-acting and short-acting antibiotics, respectively, and also in comparison to control groups without antibiotic treatment. A total of 684 cows from 288 different herds in three Norwegian regions fulfilled the criteria of the study design. There were 104 cows in control group A (sampling only), 115 cows in control group B (placebo), 221 cows treated with long-acting intramammaria Benestermycin vet. 'Leo' for 1 day at drying off in group C, and 244 cows treated with four short-acting intramammaria Leocillin with Dihydrostreptomycin vet. 'Leo' every second day before drying off in group D. The overall effect, measured as the cow being healthy after therapy, was 14.2% in control groups and 33.7% in therapy groups 30 +/- 17 days into the next lactation. Of quarters infected with S. aureus both in late lactation (45 +/- 32 days before drying off) and at drying off, 38.4% in the control group were bacteriologically negative 30 +/- 17 days into the next lactation, compared with 49.5% in the long-acting group and 68.6% in the short-acting group. Of quarters infected with Str. dysgalactiae both in late lactation (45 +/- 32 days before drying off) and at drying off, 10 out of 27 were still infected with Str. dysgalactiae in the control group 30 +/- 17 days into next lactation, compared with 0 out of 31 in the therapy groups. Dry-cow therapy in coagulase-negative Staphylococcus spp. (CNS)-infected quarters led to a 5.2 odds ratio of being healthy quarters 30 +/- 17 days into the next lactation, compared with control groups. Despite this, the overall frequency of CNS in the material was unchanged after therapy compared with controls. Short-acting compared to long-acting preparations had a significantly better effect in preventing new infection with S. aureus or Str. dysgalactiae in untreated healthy quarters in cows with fewer than three infected quarters. This difference in preventive effect was greater in cows with one infected quarter during previous lactation (the new infection rates being 0.078 for short-acting and 0.149 for long-acting) than in those with two infected quarters (the new infection rates being 0.042 and 0.063, respectively).