Evaluation of milk antibiotic residue screening tests in cattle with naturally occurring clinical mastitis

Milk from 172 commercial cows with mild to moderate clinical mastitis was tested with 5 antibiotic residue detection assay systems. 100 cows were treated with 1 of 2 intramammary β-lactam antibiotics and the remaining 72 cows were treated with intramuscular oxytocin. Milk samples were collected pre-treatment, twice after therapy and again 21 days after the initiation of treatment. Presumptive false-positive assay results were tabulated from all pre-treatment and 21-day milk samples, and from samples collected following oxytocin therapy. The percentage of false-positive results was 43.6, 37.7, 81.7, 2.6 and 18.8 for the CITE® probe (β-lactam), Delvotest-P®, Charm Farm®, LacTek® (β-lactam) and Bacillus stearothermophilus var. calidolactis disc assay respectively. In 4 of the assay systems, average somatic cell counts were significantly higher in samples yielding false-positive results than in those with negative results. Specificity and sensitivity were estimated for each assay system and, based on these estimates, positive and negative predictive value curves were plotted as the prevalence of milk samples containing detectable concentrations of exogenous antibiotic residues in the sample population varied from 0 to 100%.