Verdonk G.P.H.T.,Microbiology Control Laboratory B |
Willemse M.J.,Microbiology Control Laboratory B |
Hoefs S.G.G.,Microbiology Control Laboratory B |
Cremers G.,Microbiology Control Laboratory B |
van den Heuvel E.R.,Microbiology Control Laboratory B
Journal of Microbiological Methods | Year: 2010
Classical microbiological methods have nowadays unacceptably long cycle times. Rapid methods, available on the market for decades, are already applied within the clinical and food industry, but the implementation in pharmaceutical industry is hampered by for instance stringent regulations on validation and comparability with classical methods. Equivalence studies become less relevant when rapid methods are able to detect only one single microorganism. Directly testing this capability is currently impossible due to problems associated with preparing a spiked sample with low microbial counts. To be able to precisely estimate the limit of detection of rapid absence/presence tests, the method of the most probable limit is presented. It is based on three important elements; a relatively precise quantity of microorganisms, a non-serial dilution experiment and a statistical approach. For a set of microorganisms, a limit of detection of one was demonstrated using two different rapid methods. © 2010 Elsevier B.V.