Evaluation of point-of-care tests for identification of pathogens to inform clinical mastitis treatment decisions in pasture- and confinement-managed dairy cows in Australia.

To support antimicrobial stewardship in livestock production, there is a growing array of point-of-care diagnostics to guide antimicrobial treatment. The primary objective of this observational study was to evaluate the diagnostic performance of 5 point-of-care tests currently available in Australia for guiding lactational treatment of nonsevere clinical mastitis. A secondary objective was to describe the pathogen profiles of mastitis-causing organisms in cows managed in barns ("intensive") and on pasture ("nonintensive"). Foremilk samples (n = 641) were collected by farm staff in dairy herds in Australia (n = 30) and tested at a university laboratory using a reference test and 5 index tests. The reference test was aerobic culture on trypticase soy agar with 5% sheep blood followed by MALDI-TOF for identification of isolates. The following point-of-care tests were evaluated as index tests: Accumast, biplate, Check-Up, Mastatest (with AUP3-2022 product), and 3M Petrifilm. We found that 23% of samples were contaminated, with the median herd contamination prevalence being 22%. After excluding contaminated samples, the most common diagnoses (according to the reference test) in intensive herds were no growth (31.7%), Klebsiella spp. (28.1%), Escherichia coli (15.0%), and Streptococcus uberis (8.4%). The most common diagnoses in noncontaminated samples from cows in nonintensive herds were Strep. uberis (35.0%), no growth (26.9%), and E. coli (13.3%). After 24 h of incubation, all index tests demonstrated limited diagnostic sensitivity for identification of pathogens of interest (range: 0.06-0.63). Diagnostic performance was better at the group-level, with sensitivity and specificity for identification of noncontaminated gram-positive growths (i.e., cases that are widely considered to be candidates for antimicrobial treatment) being 0.84 and 0.75, respectively (biplate), 0.76 and 0.90 (Accumast), 0.89 and 0.79 (Check-Up), 0.67 and 0.83 (Petrifilm), and 0.55 and 0.81 (Mastatest). In intensive herds, 22.7% to 40% of cases were classified as antimicrobial treatment candidates by index tests, which was less than for cows in nonintensive herds (41.3%-61.0%). Despite limited diagnostic reliability at the genus level and species level, and the need to ensure samples are collected aseptically, our findings indicate that implementation of selective treatment protocols using the tests evaluated in this study would likely reduce antimicrobial usage in Australian herds.