ABSTRACT
This study aimed to evaluate the diagnostic performance (specificity, Sp; sensitivity, Se; accuracy; positive predictive value; negative predictive value; and Cohen's kappa coefficient, κ, of agreement) of chromogenic culture media for rapid identification of microorganisms isolated from cows with clinical (CM) and subclinical mastitis (SCM). For this, 2 experiments were carried out: evaluation of (1) biplate, and (2) triplate of chromogenic culture media for rapid identification of mastitis-causing microorganisms. For the evaluation of diagnostic performance, identification of microorganisms by MALDI-TOF mass spectrometry was considered the standard methodology. In experiment 1, 476 milk samples collected from cows with CM and 660 from cows with SCM were evaluated by inoculation in 2 selective chromogenic culture media (CHROMagar) for gram-positive bacteria and another for gram-negative bacteria. In experiment 2, 476 milk samples from cows with CM and 500 from cows with SCM were evaluated by inoculation in triplate chromogenic culture media (Smartcolor2, Onfarm), selective for Streptococcus and Strep-like organisms, Staphylococcus, and gram-negative bacteria. In experiment 1 for the CM samples, the use of biplates with gram-positive and gram-negative culture media showed Se that ranged from 0.56 (0.32-0.81; Staphylococcus aureus) to 0.90 (0.80-0.99 Streptococcus uberis), Sp varied from 0.94 (0.92-0.96; Strep. uberis) to 1.00 (Prototheca spp. or yeast), and κ ranged from 0.47 (0.26-0.67; Staph. aureus) to 0.84 (0.78-0.9; Escherichia coli). The Se of biplates for SCM samples ranged from 0.50 (0.15-0.85; E. coli) to 0.94 (0.87-1.00; Staph. aureus), Sp varied from 0.95 (0.93-0.97; Strep. uberis) to 0.99 (0.98-1.00; Staph. aureus and Strep. Agalactiae or dysgalactiae), and κ ranged from 0.18 (0.00-0.40; Escherichia coli) to 0.88 (0.80-0.95; Staph. aureus). In experiment 2, the Se of the triplate chromogenic media in CM samples ranged from 0.09 (0.00-0.26; Serratia spp.) to 0.94 (0.85-1.00; Klebsiella spp. and Enterobacter spp.), Sp varied from 0.94 (0.92-0.96; Strep. agalactiae and Strep. dysgalactiae) to 1.00 (Serratia spp.) and κ ranged from 0.07 (0.00-0.24; Serratia spp.) to 0.85 (0.75-0.94; Klebsiella spp. and Enterobacter spp.). For SCM samples, the use of the triplate with the chromogenic culture media showed Se that varied from 0.25 (0.10-0.40; Lactococcus spp.) to 1.00 (Strep. Agalactiae or dysgalactiae), Sp ranged from 0.92 (0.90-0.94; Strep. Agalactiae and Strep. dysgalactiae) to 0.99 (0.98-1.00; Klebsiella spp. and Enterobacter spp.), and κ varied from 0.28 (0.00-0.72; E. coli) to 0.72 (0.60-0.82; Staph. aureus). Our results suggest that the diagnostic accuracy of the biplate and triplate of chromogenic culture media varies according to pathogen, and the results of chromogenic culture media may be useful for rapid decision-making on mastitis treatment protocols of the main mastitis-causing microorganisms, but their use for implementation of mastitis control measures will depend on each farm specific needs.