Abstract
Antimicrobial resistance necessitates more judicious antimicrobial use, including in dairy production where mastitis remains the primary driver of antibiotic consumption. Selective treatment of nonsevere clinical mastitis (STCM) requires rapid differentiation between gram-positive and gram-negative pathogens, as only gram-positive infections generally benefit from antimicrobial therapy. Because conventional laboratory culture is too slow for STCM, rapid testing systems for pathogen detection in milk are used on farm (performed and analyzed by farm personnel) or on practice (performed and analyzed at the veterinary practice). This study evaluated the diagnostic performance of 2 culture-based rapid tests (MicroMast and VetoRapid) via an on-practice culture approach as part of a clinical trial, with standard laboratory bacteriology as the gold standard. The performance to distinguish (1) gram-positive from gram-negative bacteria and (2) further pathogen identification was assessed by determining sensitivity, specificity, positive predictive values, negative predictive values, accuracy, and Cohen's kappa values. Both tests (VetoRapid more than MicroMast) reliably distinguished gram-positive from gram-negative pathogens, supporting their use in STCM decision-making. However, their ability to identify specific pathogens was limited, restricting their suitability for broader udder health management decisions.