A single link to the first track to allow the export script to build the search page
  • President’s Report
  • Validation of Real-time PCR and Bacteriological Culture for Identification of Streptococcus agalactiae and Staphylococcus aureus in Milk and on Teat Skin in Herds with Automatic Milking System

    • Nanna K. Skjølstrup ;
    • Louise R. Mathiasen ;
    • Ilka C. Klaas ;
    • Line Svennesen ;
    • Yasser S. Mahmmod ;
    • Karl Pedersen
    Please login to access the content.

    Validation of Real-time PCR and Bacteriological Culture for Identification of Streptococcus agalactiae and Staphylococcus aureus in Milk and on Teat Skin in Herds with Automatic Milking System 
    Nanna K. Skjstrup1, Louise R. Mathiasen1, Ilka C. Klaas2, Line Svennesen1, Yasser S. Mahmmod1, Karl Pedersen3 1University of Copenhagen, Frederiksberg C, Denmark 2DeLaval International AB, Tumba, Sweden 3Technical University of Denmark, Kongens Lyngby, Denmark 
    Despite a national surveillance program for Streptococcus agalactiae, an increasing herd-level prevalence is seen and most Danish dairy herds are already infected with Staphylococcus aureus (Katholm et al. 2012). Intramammary infections (IMI) with these pathogens result in financial losses for farmers and impaired animal welfare for cows. Even though both bacteria are categorized as contagious mastitis pathogens, environmental reservoirs are described in scientific literature (Jorgensen et al. 2016). The importance of these environmental reservoirs is still discussed and needs further investigation, especially for S. agalactiae. Valid diagnostic tests are essential for correct diagnosis and treatment. This study’s objective was to estimate sensitivity (Se) and specificity (Sp) for real-time PCR assay ‘Mastit4’ against bacteriological culture (BC) for identifying S. agalactiae and S. aureus in milk and on teat skin.  
    Materials and Methods The study involved eight herds, which were repeatedly positive for S. agalactiae in bulk tank milk in spring 2017 and operated with automatic milking system. In each herd, a random sample of 30-40 cows with high somatic cell count (SCC =200,000 cells/mL) was selected. The study design was based on two parts according to sampling technique. In part A, teat skin samples were collected using the wet-dry method described by Paduch and Krker (2011) and aseptic milk samples from all four quarters were collected for BC, whereas only right rear (RR) quarters were selected for both BC and PCR analysis. Selected positive quarters were resampled with PCR and BC (part B) one to three weeks after first sampling (part A) where PCR samples were taken with FLOQ swabs. PCR samples were analyzed at the analysis laboratory of DNA Diagnostic A/S (Aarhus, Denmark). Milk and teat skin samples were cultured on calf blood agar, modified Edward’s medium and SASelectTM medium (selective for S. aureus). For milk samples, 10 µl was streaked onto a quarter of an agar plate, whereas for teat skin samples, 100 µl was inoculated on a whole agar plate and spread with a drigalski spatula. In part B, no SASelectTM medium was used. All suspected colonies were confirmed using MALDI-TOF (S. aureus) or latex agglutination test (S. agalactiae). In part A, PCR swabs were immersed into sample tubes after streaking (day after sampling), whereas in part B, PCR swabs were rolled directly on teat skin or immersed into fresh milk samples, respectively. In part A and B, Se and Sp of PCR and BC for S. agalactiae and S. aureus in milk were estimated using latent class analysis (LCA), whereas Se and Sp for PCR of S. agalactiae and S. aureus on teat skin were estimated relative to BC. Also, agreement (kappa) between PCR and BC was estimated.  
    Results In part A, 287 RR quarters were analyzed with BC and PCR, in part B this number was 132 quarters. Using LCA Se and Sp of BC and PCR for the diagnosis of IMI with either S. aureus or 
    S. agalactiae in both part A and B were estimated. For S. agalactiae in milk in part A, Se and Sp of PCR were 96.4 and 93.4 %, respectively, whereas Se and Sp of BC were 82.4 and 99.7 %, respectively (kappa = 0.61). For S. aureus in milk in part A, Se and Sp of PCR were 87.6 and 
    98.2 %, respectively, whereas Se and Sp of BC were 74.0 and 99.4 %, respectively (kappa = 0.72). For S. agalactiae in milk in part B, Se and Sp of PCR were 95.4 and 91.4 %, respectively, whereas Se and Sp of BC were 88.6 and 97.9 %, respectively (kappa = 0.74). For S. aureus in milk in part B, Se and Sp of PCR were 93.2 and 94.7 %, respectively, whereas Se and Sp of BC were 57.7 and 99.4 %, respectively (kappa = 0.58). Due to a low number of teat skin samples positive for S. agalactiae or S. aureus with BC, LCA was not applicable. Therefore, Se and Sp were calculated using BC as a reference standard. Compared with BC, Se and Sp of PCR on teat skin were 100 and 82.2 %, respectively, for the identification of S. agalactiae in part A (kappa = 
    0.031) and 30.4 and 86.7 %, respectively, for the identification of S. aureus in part A (kappa = 0.13). Compared with BC, Se and Sp of PCR on teat skin were 100 and 35.9 %, respectively, for the identification of S. agalactiae in part B (kappa = 0.0008) and 90.9 and 42.1 %, respectively, for the identification of S. aureus in part B (kappa = 0.086). 
    Discussion SePCR was higher than SeBC for S. aureus and S. agalactiae in milk samples in both part A and B. This indicates that PCR analysis can be carried out with the FLOQ swab method as alternative to analyze milk. The differences in test properties of PCR and BC can be explained by different factors, e.g. variation in bacterial shedding, sample inoculum, the underlying disease definition, the presence of DNA in non-viable bacteria, and limitations of BC due to differences in bacterial growth and laboratory methods chosen. The poor agreement between PCR and BC on teat skin indicates that the two methods do not measure the same: PCR detects viable as well as dead and inactivated bacteria whereas BC detects only viable bacteria. Bacteria may have been inactivated by disinfectant teat spray, which was applied after milking.  
    Conclusions The agreement between BC and PCR for detecting S. agalactiae and S. aureus on teat skin was poor, thereby indicating that the two methods do not measure the same condition. Further studies are needed to investigate the microbiota of the teat skin. Sensitivity for diagnosing IMI caused by either S. agalactiae or S. aureus was higher for PCR than BC. The agreement between PCR and BC on milk is good, indicating that PCR is suitable for milk samples with or without applying the FLOQ swab method. The use of PCR for diagnosing IMI is therefore recommended. 
    This study was funded by the Danish Milk Levy Fund and DNA Diagnostic A/S. 
    References Jorgensen, H.J., A.B. Nordstoga, S. Sviland, R.N. Zadoks, L. Sver, B. Kvitle, and T. Mk. 2016. Streptococcus agalactiae in the environment of bovine dairy herds – rewriting the textbooks? Vet Microbiol. 184, pp 64–72. 
    Katholm, J., T.W. Bennedsgaard, M.T. Koskinen, and E. Rattenborg. 2012. Quality of bulk tank milk samples from Danish dairy herds based on real-time polymerase chain reaction identification of mastitis pathogens. J. Dairy Sci. 95:10, pp 5702–5708.