MIC analysis - (Jun/18/2014 )
I am after some advice on analysing MIC data. We are testing the efficacy of a disinfectant and started with MIC to establish the working concentration of the product. This was done by carrying out a two fold dilution in a 96 well microtiter plate. This was repeated several times for each organism. I appreciate that the MIC is the lowest dilution that inhibits growth however as this test was replicated we have some bacteria such as Staphs that have different MICs on each plate! This data was from the same strain. Some colleagues have suggested taking the average and carrying out stats to get the range at 99% CI however as this is a two fold dilution it is obviously quite large as the concentration of disinfectant in each well is halved each time. Can averages be taken for MICs like this? What stats can I use to analyse this data?
MIC has limited relevance to disinfectant efficacy. Whereas you shouldn't see much variability in MIC for the same isolate, your wasting your time with extensive statistical analysis of such data if your focus is only disinfectancy. Suggest you look up the AOAC use-dilution and EN suspension methodologies.
In what context do you project application of your work (e.g. household, hospital)? This will direct to organisms against which your would evaluate efficacy.
Hi Phil Geis
Thanks for your response.
The company making the disinfectant is relatively new and want their product in a ranch of applications, because of this we are testing it against various organisms just to see how it performs. We will be following the EU standards for disinfectant efficacy including suspension tests at some point in the future, however these test require us to know the working concentration in order to do the suspension methodology. As we currently do not know this, we did MICs. The MIC results show the disinfectant works but my main problem is interpreting this information. For example the same strain of bacteria show MICs with quite a big range in replicated tests (0.05-0.2% disinfectant). I do not know if I can just average these results from my replicates and use this value as a working concentration, especially as a two fold dilution was used in the MIC. Basically I am trying to get the percentage of disinfectant to use and range which is statistically significant from my MIC data, however I have never done anything like this before and I do not know if it can be done.
Any help or advice would be very much appreciated.
Working concentration would be based on MBC rather than MIC. Bactericidal efficacy (vs. stasis - MIC) is the effective expectation of disinfectancy (by EN, 4 or 5 log reduction by 5 minutes) so MIC per se does not show that a disinfectant works..
Be aware that efficacy of some disinfectants (iodophors, quats) may not be a simple function of dilution - greater concentrations can be less effective. However concentrations mentioned don;t appear to fall within the ranges for such phenomena. Could you have pH, phase or solubility issues with the product as its diiluted? Strongly suggest you use a suspension MBC protocol. Do you anticipate testing of product with EN 1040, 1276, 1397, other?
Do you get consistent results at the same concentration? In any case, I'd not average results - MIC (and MBC) should be specific and reproducible endpoints.
Hi Phil Geis
Thanks again for your response, I really appreciate your advice.
I took your advice and did MBCs and just read the results today. As to be expected the bactericidal efficacy of the disinfectant showed higher concentrations than the MICs however I am still getting inconsistent results with my replicates. Basically, to answer your question I am not getting consistent results at the same concentration. You could be right about having pH, phase or solubility issues with the product as it is diluted, although I made sure everything was mixed thoroughly during every step of the MBC. Other than that I have not got a clue why I am getting different results with the same strains.
I will be using EN 1040 and 1276 in the future however as I currently have 5 different species of bacteria with 5 strains in each, all showing inconsistent results on the MBC replicates I cannot see me doing this any time soon.
Sure wish i could help - it's hard to understand that apparent replicate testing of the same dilution and isolate give inconsistent results.