Browsing by Author "Shinga, Mduduzi Hendrick."

Now showing 1 - 1 of 1

Investigating alternative methods to detect bovine mastitis in milk.
(2018) Shinga, Mduduzi Hendrick.; Laing, Mark Delmege.; Basdew, Iona Hershna.
The aim of this study was to investigate alternate measures for the diagnosis of bovine mastitis, which can either be done separately or coupled with the current use of somatic cell counts. Techniques that were assessed include measurement of pH, electrical conductivity (EC) and volatile organic compounds (VOCs) liberated by pathogens during metabolism in milk; the quantification of milk components (fats, whey proteins, lactose, caseins), and cell counts of Staphylococcus aureus. Various concentrations of S. aureus were used to assess the minimum bacterial inoculum level that could bring about detectable changes in the pH and EC values of milk. It was found that 10-2 diluted inoculum caused less changes in pH and EC compared to the undiluted bacterial inoculum size. On average, the pH of milk samples decreased from 6.45 to 5.31 after 32 hours of incubation at 37˚C. A corresponding EC value increased from 5.28 mS cm-1 to 6.68 mS cm-1 was observed due to the liberation of sodium and chloride ions during the incubation of inoculated milk after an incubation of 32 hours. The detected VOCs including compounds from hydrocarbon, ester, ketone, aldehyde and siloxane groups were observed from milk inoculated with common mastitis pathogens such as Staphylococcus aureus, Streptococcus uberis, Streptococcus agalactiae, coagulase negative staphylococci (CNS) and Escherichia coli. Only 20% of a total of 50 inoculated samples released VOC’s. Furthermore, the VOCs identified were not species-specific. However, by comparing the samples to a control (un-inoculated sample), the identified VOCs could be used as a rough monitoring tool to distinguish inoculated milk from un-inoculated milk. Near-infrared analysis (NIRA) was carried out using the Kernel partial least squares regression. However, the calibration models for milk composition and S. aureus were poor. We believe that this was affected by the technique used, measuring the NIR absorbance of milk samples in plastic Petri dishes. The absorptive abilities of polystyrene present in Petri dishes affected the NIRS scans. Secondly, insensitive wet chemistry methods, and the low sample number used in this study were concluded to be the major reasons for the poor predictive models that were obtained for the analysis of milk components and S. aureus. These analytic tools showed potential as diagnostic methods, however, further research must be conducted to solve these problems.