Implementing considered elements of standardisation for Time Kill Curve experiments across multiple sites: A European collaboration perspective
Attwood, M. L. G.; Bronstrup, M.; Das, S.; Fuchs, H.; Griffin, P.; Hinkelmann, B.; Hoare, L.; Lebrat, J.; Marchand, S.; Mercer, D.; Michel, F.; Noel, A.; Nussbaumer-Proll, A.; Zeitlinger, M.; MacGowan, A. P.
Show abstract
SynopsisO_ST_ABSBackgroundC_ST_ABSThe main advantages of Time Kill Curves (TKCs) in antimicrobial drug development are the ability to track bacterial kill and regrowth over time and with varying drug concentrations. Whilst there are guideline documents in place, such as M26-A in CLSI, there remains scope for individual laboratory differences in practice. Here we evaluated several factors which potentially influenced data generated in TKCs. MethodsFirstly, E. coli ATCC 25922 was used to determine optimum sampling volume, culture vessel volume, CFU enumeration variance factors and static versus agitated cultures in a single laboratory. Secondly, a ring test comprising of TKCs was performed by six laboratories focusing on: standardised inoculum, static culture and two culture vessel sizes 10 mL and 200 {micro}L. Data analysis was performed to determine consistency within centres and between them. ResultsConsistently accurate inocula could be achieved by use of: larger sampling volumes between 100 {micro}L > 20 mL; larger culture vessels volumes (10 mL > 100 {micro}L) and higher inocula (10 8 > 1.5x10 5 CFU). Culture agitation during the TKC experiment resulted in reduced killing compared to static cultures. Reproducibility of TKCs was best between centres when they were performed in 10 mL culture vessels. There was more variability per site when performing TKC in 96 well trays. ConclusionsTechnical factors such as preparation of inocula, agitation, vessel size and enumeration of cultures are important variables in performing TKCs that need to be standardised in drug development programmes involving multiple laboratory centres.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.