My laboratories still perform anti-microbial susceptibility testing by the “old fashioned” disc diffusion methodology. We have never been able to justify the costs of automated susceptibility testing (AST)systems such as Vitek or Phoenix, and the cost differences are not insignificant. (approx $1 v $8-10 per test). Multiply that by a few thousand tests and you see what I mean.
The clinicians seem to be happy with this as are the accreditation agencies. I appreciate that the turnaround time might be a bit quicker for AST systems, but is it worth the cost. I am not sure…
AST systems such as Vitek and Phoenix also give you an MIC approximation. However I think zone diameter also gives you a good indicator of how susceptible or how resistant an organism may be in-vitro, which is clearly important in some clinical cases.
I think the current AST systems promote susceptibility testing that is unnecessary, both in terms of the number of antibiotics that are tested against, and also the number of isolates that are set up for susceptibility testing. This adds to the cost further, and can also promote unnecessary use of antibiotics.
It is interesting that the TLA systems that are being developed at the moment are looking to incorporate both disc diffusion (plating, disc placing, reading & interpretation) and AST systems within their processes. This suggests to me that both approaches will be around for many years yet.
I don’t believe that our laboratories are inferior because we don’t have a Vitek or a Phoenix or something similar . On the contrary the use of disc diffusion makes us think hard about what isolates we submit for susceptibility testing, and we are very selective regarding the antibiotics we test against, As a result of this a smart, efficient and clinically acceptable approach to susceptibility testing has evolved within our laboratory network.
As much as I am a fan of automation, I like the considered approach to susceptibility testing that disc diffusion offers, and would require some degree of convincing to move away from it….
Michael
Here in the U.S. turnaround time is everything because of the way insurance pays hospitals. The faster a doctor gets a result the faster they can get treated and the sooner the patient can hopefully be discharged, many times to their home with IV antibiotics taken care of by visiting nurse services. I cannot imagine going back to the tediousness of reading Kirby Bauer sensitivities with the volume of work we do. We process around 150 urines a day M-F and read plates on two shifts with a 24 hour incubation time for no-growths. Sensis that day shift sets up on Vitek2 are read out by second shift. Sensis that 2nd sets up are read out by day shift. We resort to KB for some organisms or discrepancies. It’s rare that I set up more than 1-2 per week. We are totally paperless, and the sensitivities cross over with a bi-directional interface between the LIS and Vitek2. We “bring over” the sensi once it crosses the interface, check it, and sign it out in about 15 seconds. Labor is costly too. Add up all the time you spend reading zone sizes, recording them, and signing out a culture. I can set up an E. coli for the Vitek2 in less than a minute, and report it in less than a minute. It takes me about a minute to put in workups and do indole and oxidase and I’m done.
As far as antibiotics that are reported out, certain ones are on the sensi cards but the results are suppressed. Infectious disease doctors are usually the ones that want something un-suppressed and we can go into the culture and un-suppress the result and release it in about a minute. The antibiotics we run are chosen by the infection control committee. I believe this is made up of micro management, pathology, pharmacy, and infectious disease doctor representatives.
Infectious disease doctors want MIC results. If Vancomycin gives an MIC of >=2, we automatically set up Ceftaroline. ID is usually on the case by then anyway. You can’t do that with disc diffusion.
Thanks for your comment Judy, I think it may be a case of what you are used to. This also applies to ID doctors. If they are used to getting MIC results, they will want MIC results! It will be interesting to see in what direction these incoming TLA systems turn towards in terms of disc diffusion or broth based susceptibility testing. I suspect both will be incorporated to allow for different demands and budgets.
One of the larger hospitals in this area, the University of Michigan System (everything is a “system” here now) in Ann Arbor, MI, finally ditched Kirby Bauer testing about 5 years ago. They are a very high-volume laboratory. From the gossip I’ve heard, the reason is that their manager retired and the new manager went to automated sensis.
Be careful you don’t get caught being “dated.” When I first started in micro in 1974, the Ph.D. in charge of our department still used tubed media for identifications. When the API system came out, he was loathe to change, and it took him a few years to make the switch. However, when the first Bactec blood culture system came out (with bottles that had to be read individually by hand and used a radioactive marker in them) we were one of the first labs in the country to be on board with it. I think it depends greatly on the population you serve, payment methods, and what your doctors demand.
We have a TLA (Kiestra) system going into our laboratory later on this year, so I don’t think we are too dated! With regards to the above issue, I really have no idea in what direction the TLA will take us. Will be interesting to see.
Many thanks for your comments.
Michael