The importance of a correctly made antibiogram cannot be overstated. It is the very basis for evidence based antimicrobial prescribing. It should conform to the CLSI/EUCAST guidelines. Once an antibiogram is ready, we should examine it critically and identify flaws and rectify them.
Some Red Flags to watch out for:
1. Amikacin percentage sensitivity is lower than gentamicin.
2. Ceftriaxone/Cefotaxime/ Ceftazidime percentage sensitivity is lower than cefuroxime.
3. Cefuroxime percentage sensitivity is lower than cefazolin
4. Piperacillin tazobactam percentage sensitivity is lower than or equal to amoxicillin-clavulanic acid and/or Ceftriaxone/Cefotaxime/ Ceftazidime
5. Meropenem percentage sensitivity is same or lower than ertapenem and/or piperacillin tazobactam.
6. Ciprofloxacin percentage sensitivity is higher than levofloxacin
In all these instances, there may be a mistake with the denominator. Please make sure the assessment of sensitivity is based on numbers tested.
Making a correct representative antibiogram is serious business.
It is the foundation of judicious antimicrobial prescribing.
It should then be shared with the Clinical Colleagues. Best way of communication is presenting it during their seminar. This allows brainstorming and consensus building and engagement with them.
#Issue of calculating antimicrobial susceptibilities if two panels are being tested. How to avoid incorrect calculations:
Case in study: meropenem versus piperacillin tazobactam in E. coli
Let the total number of isolates tested be 100. It was found that 20 isolates were resistant to the first panel which contained say gentamicin, cefuroxime, cefotaxime, ceftriaxone, ceftazidime, ciprofloxacin, augmentin and piperacillin tazobactam. These 20 were tested by second panel which will include meropenem.
When calculating susceptibility of the drugs used in the second panel, an error will occur if the numerator and denominator are extracted from only those isolates tested by this panel. It will definitely give a skewed result. e.g meropenem sensitivity may be lower than expected e.g : 78-82%
This is because only strains resistant to the first line (say 20) were tested by the second panel. An error will occur if only the results of these 20 isolates are used to assess sensitivity to meropenem- It will be skewed for greater resistance. Say 12 out of 20 were susceptible to meropenem. So one can mistakenly say susceptibility of meropenem is 12/20 x 100= 60%. THIS IS AN EASY TRAP as also commented by Dr David Livermore. Here Numerator and denominator were extracted from the second panel.
However, it is important to understand that meropenem is bound to be sensitive in all the 80 isolates which were tested only by the first panel as it is largely resistant to hydrolysis by ESBLs and AmpC except if there is an occasional OXA-48- that is why it is in the second panel!
So, the numerator for meropenem should be 80+12=92
Denominator remains 100
Susceptibility of meropenem= 92/100 x100%=92%
Now 80 isolates were sensitive to piperacillin tazobactam by the first panel.
It’s sensitivity remains 80/100 x100= 80%
I hope this clarifies the issue.
Nitrofurantoin for Pseudomonas/ Proteus/ Enterobacter. When calculating overall sensitivity of nitrofurantoin, please omit these from both numerator and denominator.
Ampicillin for Klebsiella/Morganella/ Citrobacter/ Enterobacter/ Serratia
Cotrimoxazole for Pseudomonas and Enterococci
First/ Second and even third generation cephalosporins for Enterobacter cloacae, Citrobacter freundii, Serratia marcescens as they have inducible AmpC and the patient will develop resistance while on treatment.
Fosfomycin for Staph saprophyticus/ Acinetobacter
Amoxicillin-clavulanic acid/cefaclor/ceftriaxone/cefotaxime for Pseudomonas aeruginosa
The primary aim of preparing a local antibiogram is to guide clinicians in optimum selection of empirical antimicrobial therapy for initial infections. In uncomplicated UTI it has even more significance because the treatment is almost always empirical.
The antibiogram shows the trend of antimicrobial susceptibility over the last one year and thus efficiently guides therapy.
The recommendations as per CLSI for developing an appropriate antibiogram are as follows:
1. Targeted, focused antibiograms should be prepared
2. Only final, verified test results will be included.
3. Only species with testing data for more than 30 isolates will be analysed and included in the antibiogram. Each centre will focus on the top five species identified in their centre, which are usually Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis, Pseudomonas aerginosa, Acinetobacter baumanii complex, Staphylococcus saprophyticus Enterococcus faecalis. If a centre has more than 30 isolates of the following, then they too may be included: Citrobacter koseri, C. freundii, Enterobacter cloacae, Klebsiella aerogenes, Serratia marcescens
3. Only diagnostic isolates from samples coming from the OPD or Emergency Department will be included in the first year.
4. Only the first isolate of a species/patient/analysis period will be included irrespective of antimicrobial susceptibility profile.
5. Only routinely tested antimicrobial agents will be included to calculate the percent susceptible from results reported. If restricted formulary is being used, include both released and suppressed results. Supplemental agents selectively tested on resistant isolates only should not be reported.
6. When reporting the percentage susceptible do not include the percent intermediate in the statistic.
Consent from participating patients:
Consent from individual patients will not be needed as we are not taking any confidential information. Consent form is not required for this study. For this study protocol, a waiver of consent has been proposed. There is little conceivable risk to participants as this is a non-interventional study and is using observational data only that is routinely collected. There are also no anticipated adverse effects for participants given that data collected is done so in a confidential manner and is observational data only.
Cotrimoxazole for Acinetobacter/Burkholderia/Stenotrophomonas
Ceftazidime for Acinetobacter/Burkholderia/Stenotrophomonas
When should Immediate Action be taken while preparing a report?
When a MDR/XDR/PDR strain is isolated. Immediately contact the Physician/ Surgeon and check the veracity of the sample. What is the patient’s status? If stable, then treat it as a contaminant/commensal. One can request a fresh well collected sample if in doubt. If the isolate is treated as commensal it should be indicated as such when being documented. Such isolates should be excluded when building the antibiogram. Otherwise we are creating a wrong antibiogram which is skewed towards resistance and by sharing/publishing it we are driving the resistance.
The importance of a correctly made antibiogram cannot be overstated. It is the very basis for evidence based antimicrobial prescribing. It should conform to the CLSI/EUCAST guidelines. Once an antibiogram is ready, we should examine it critically and identify flaws and rectify them.
1. Amikacin percentage sensitivity is lower than gentamicin.
2. Ceftriaxone/Cefotaxime/ Ceftazidime percentage sensitivity is lower than cefuroxime.
3. Cefuroxime percentage sensitivity is lower than cefazolin
4. Piperacillin tazobactam percentage sensitivity is lower than or equal to amoxicillin-clavulanic acid and/or Ceftriaxone/Cefotaxime/ Ceftazidime
5. Meropenem percentage sensitivity is same or lower than ertapenem and/or piperacillin tazobactam.
6. Ciprofloxacin percentage sensitivity is higher than levofloxacin
In all these instances, there may be a mistake with the denominator. Please make sure the assessment of sensitivity is based on numbers tested.
Making a correct representative antibiogram is serious business.
It is the foundation of judicious antimicrobial prescribing.
It should then be shared with the Clinical Colleagues. Best way of communication is presenting it during their seminar. This allows brainstorming and consensus building and engagement with them.