Culture tests guide selective antibiotic use in dairy cow mastitis cases

Clinical mastitis remains the main reason for antimicrobial use on dairy farms. Yet a share of cases does not benefit from antibiotics: infections by gram-negative bacteria usually resolve on their own, and samples with no bacterial growth have nothing to treat. Rapidly separating those cases from the rest is the basis of so-called selective treatment, and that is where rapid culture tests run on the farm or at the veterinary practice come in.

A study published in June 2026 in JDS Communications, a journal of the American Dairy Science Association, evaluated two such kits, MicroMast and VetoRapid, within a clinical trial conducted in Belgium. The tests were performed and read at the veterinary practice, and results were compared with conventional laboratory bacteriology as the reference standard. The authors calculated sensitivity, specificity, predictive values, accuracy and agreement (Cohen's kappa) for two tasks: distinguishing gram-positive from gram-negative bacteria, and identifying the pathogen.

In the first task, both tests reliably separated gram-positive from gram-negative bacteria, with VetoRapid performing better. According to the authors, this supports using the kits when deciding whether to treat a non-severe case of clinical mastitis. In the second task, identifying the specific pathogen, the tests' ability was limited, which restricts their use for broader udder-health management decisions such as choosing strategies against contagious pathogens.

The available abstract does not report the numerical values of sensitivity and specificity; reading the full open-access article is necessary to assess the magnitude of those measures before transposing them to another production setting. The study was also conducted in a single country within a clinical trial, which limits generalisation.

An invited review published in 2023 in the Journal of Dairy Science organised the evidence on selective treatment. According to the authors, advances of the past decade made it possible to exclude from antimicrobial treatment non-severe cases with a high probability of spontaneous cure (no bacterial cause or gram-negative, excluding Klebsiella infections) and chronic cases with a low bacteriological cure rate. The review concludes that relatively fast identification of the causative agent is the most important factor in the protocol and that many studies found no detrimental udder-health consequences after adopting selective treatment with on-farm testing. The reduction in antimicrobial use, in turn, depended on the pathogen distribution in each herd and on protocol characteristics.

A 2024 Australian observational study, also in the Journal of Dairy Science, adds a note of caution about the sampling step. Among 641 milk samples collected by staff on 30 farms, 23% were contaminated, and the five point-of-care tests evaluated showed limited sensitivity for identifying pathogens of interest after 24 hours of incubation. Pathogen profiles also differed between barn-managed herds, where no-growth samples and Klebsiella spp. predominated, and pasture-managed herds, where Streptococcus uberis was the most common finding.

The three papers converge on one point: the value of a rapid test lies in answering a simple question, gram-positive or gram-negative, not in replacing full bacteriology. They also converge on the importance of sample quality and of somatic cell count and cow history records, elements the 2023 review builds into its decision protocol. Choosing a kit and defining what to do with each result depend on the herd's pathogen profile, the capacity to run the test correctly and the supervision of the responsible veterinarian.

For those working in dairy production, the message of the Belgian study is less a recommendation than a boundary: the kits evaluated serve gram-positive/gram-negative screening within a selective treatment protocol, but do not on their own support decisions that require knowing which pathogen is present. Before adopting any protocol, it is worth reading the full articles, checking performance under conditions similar to the herd's, and discussing implementation with the herd veterinarian.