Arizona Department of Health Services

Antibiogram Toolkit
Arizona Healthcare-Associated Infections (HAI) Program
 (This page is left intentional blank)

ii Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

A Note to Our Readers:
The objectives of the Antimicrobial Stewardship Programs (ASP) Subcommittee
are directed at education, presentation, and identification of resources for
clinicians to create toolkits of strategies that will assist clinicians with
understanding, implementing, measuring, and maintaining antimicrobial
stewardship programs.

ASP Subcommittee is a multidisciplinary committee representing various

healthcare disciplines working to define and provide guidance for establishing
and maintaining antimicrobial stewardship programs within acute care and long-
term care institutions and in the community.

Their work was guided by the best available evidence at the time although the
subject matter encompassed over one hundred references. Accordingly, the
Subcommittee selectively used examples from the published literature to provide
guidance and evidenced-based criteria regarding optimizing use of the annual
cumulative antibiogram and applications for antimicrobial stewardship programs.
The Antibiogram Toolkit reflects consensus on criteria which the Healthcare-
Associated Infections (HAI) Advisory Committee deems to represent best
practices in the interpretation and utilization of antibiogram data.

The Antibiogram Toolkit was developed by the ASP Subcommittee of the HAI
Advisory Committee in 2012-2013. This toolkit should be used in conjunction
with the guidance provided by the Clinical and Laboratory Standards Institute
(CLSI) M39-A3 consensus document entitled “Analysis and Presentation of
Cumulative Antimicrobial Susceptibility Test Data.”

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit iii

 Introduction
The Clinical and Laboratory Standards Institute (CLSI), formerly the National
Committee for Clinical Laboratory Standards (NCCLS), has published a series
of guidelines beginning in 2002 to assist in the preparation of cumulative
antibiograms. CLSI’s M39-A3 consensus document (third approved version,
2/5/2009), entitled “Analysis and Presentation of Cumulative Antimicrobial
ANTIBIOGRAMS Susceptibility Test Data”, provides guidance to clinical laboratories in the
Developing collection of data for preparation of a cumulative antibiogram. The guidelines
emphasize presenting susceptibility data in an accurate, reliable, consistent,
Cumulative
transparent, and timely manner, distributing the antibiogram to clinicians and
Susceptibility others who need access to the information, and presenting the results in a
Reports for Your manner that facilitates comparisons between health care institutions. The most
Clinicians and frequent use of a cumulative antibiogram report is to guide initial empirical
Ensuring Their antimicrobial therapy decisions for the management of infections in patients for
whom definitive microbiological susceptibilities are unavailable for the infecting
Appropriate
pathogen(s).
Interpretation and
Effective Use Most antimicrobial choices are empiric and made before the identification and
susceptibility pattern of the infecting pathogen is known. Empiric antimicrobial
choice is guided by many considerations, but local antimicrobial susceptibility
patterns of commonly isolated bacteria are paramount among them. Since
antimicrobial resistance has increased steadily in many institutions, and since
resistance rates vary by geographic location and patient demographics, the
ready availability of up-to-date cumulative antimicrobial susceptibility data is
crucial. These data are also essential to monitor emerging trends in resistance
at the local level to support clinical decision-making, evaluate infection-control
interventions and antimicrobial-resistance containment strategies, optimize
microbiology susceptibility testing and reporting methods, and guide Pharmacy
and Therapeutics Committee formulary decisions. Other applications for the
analysis of susceptibility test data may include methods not included in the
CLSI M39-A3 manual, such as identifying isolates with specific antimicrobial
resistance phenotypes.

However, cumulative antibiogram reports have significant limitations. These

should be noted as part of any educational program concerning antibiogram
use. At the same time, these limitations provide opportunities for innovation
and discussion with clinicians and infectious diseases physicians on how to
incorporate this data into future antibiogram editions. For example, a hospital
antibiogram may be less valuable when selecting subsequent therapy for a
patient with an early re-emerging infection or persistent infection because the
antibiogram uses the mean susceptibilities of a population to predict clinical
response without regard to previous antibiotic exposures in a specific patient.
Antibiograms provide susceptibility data but do not reveal additional information
concerning microbial isolates, such as the timing of the isolate in relation to
the patient’s hospital admission (i.e., to determine whether an infection was

iv Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

community or health care acquired) or patient demographics and previous antibiotic exposure. Also, antibiograms
reveal qualitative measures of susceptibility (i.e., whether a pathogen is resistant or susceptible) but do not
provide quantitative data, such as minimum inhibitory concentrations (MICs), and thereby cannot detect
significant elevations in MICs within a susceptible range which might signal acquired mechanisms of resistance
(e.g., “MIC creep”). A further limitation of antibiograms is that they only capture the aggregate proportion of
susceptible isolates for a given organism-antibiotic combination and do not provide data on the proportion of
other antibiotics that are also active (i.e., cross-resistance to multiple antibiotics). Therefore, the cumulative
antibiogram report should be viewed as a compilation of data which provides both opportunities and challenges.
By its inherent nature, antibiograms provide valuable information which is vast but at the same time limited and is
easy to misinterpret. Therefore, an active and continuous educational program is necessary.

The “Antibiogram Toolkit” is supported by the Healthcare-Associated Infections (HAI) Advisory Committee
to provide additional direction for clinicians involved in constructing the cumulative antibiogram report and
educating clinicians on it. This toolkit should be used in conjunction with approved CLSI documents and
additional literature regarding microbial resistance. The toolkit hopefully enriches discussions on the challenges
and opportunities with susceptibility data reporting. While the specific scenarios are detailed a multidisciplinary
antibiotic stewardship team should find ways to implement some of these projects and further analyze their own
antibiogram data to produce more accurate and fruitful educational activities.

The Antibiogram Toolkit contains two major components: antibiogram templates (Part I) and a set of suggestions
for constructing antibiograms, such as CLSI-defined rules, including antibiogram concepts, opportunities for
education, and antibiogram projects (Part II). Part II includes several topics which should enhance the accuracy
and utility of the cumulative susceptibility report. These were selected by the authors from professional
experience while involved with antibiotic stewardship programs during their careers. However, many more
examples can be identified from the literature. A short list of references is supplied at the end but falls short of the
dozens of examples published in the peer-reviewed literature. It is hoped that additional examples can be added in
the future.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit v

Toolkit Contents
The major recommendations for preparation, education, and solutions for common problems are provided
in the first 3 topics, as follows:
• Recommendations for preparation of a cumulative antibiogram....................................................................2

• Ideas on educating prescribers on reading and interpreting the cumulative antibiogram..............................3

• Antibiogram pitfalls and how to correct them................................................................................................4

The first three topics are followed by a series of scenarios and examples which build on solutions and
further enhance the utility of antibiograms, education of prescribers, and identification of projects.
• Contributions to antibiotic resistance may be out of your control: the importance of patient
location on susceptibilities..............................................................................................................................5

• The problem with antibiograms: numbers represent single-drug resistance..................................................7

• Detecting excessive influence of repeat (duplicate) isolates..........................................................................9

• Cascade (selective) testing and reporting: a pitfall...................................................................................... 11

• When antibiogram data fails to provide direction to narrowing the antibiotic spectrum in
select patient circumstances..........................................................................................................................12

• Presenting multi-institutional cumulative antibiogram data: local, regional, and national results...............14

• Presenting trends in resistance as an educational section of the antibiogram..............................................16

• Institutional antibiograms do not provide information on how antibiotic use is epidemiologically

linked to resistance rates...............................................................................................................................17

• Assessing resistance trends: utilizing statistical analysis to evaluate changes in susceptibility rates..........18

• Using the antibiogram as part of antimicrobial stewardship initiatives........................................................19

A short but essential reading list is provided specifically aimed at antibiograms and their use as part
of an antibiotic stewardship program.........................................................................................................................21

Appendix The cumulative antibiogram report: Templates and suggestions..............................................................22

 Recommendations for Preparation, Education, and Solutions

Recommendations for Preparation of a Cumulative Antibiogram

• Analyze/present cumulative antibiogram report at least annually.

• Include only final, verified test results.

• Include only species with testing data for ≥ 30 isolates.

• Include only diagnostic (not surveillance) isolates.

• Eliminate duplicates by including only the first isolate of a species/patient/analysis period, irrespective of
body site or antimicrobial profile.

• Include only antimicrobial agents routinely tested; do not report supplemental agents selectively tested on
resistant isolates only.

• Report percent susceptible (%S) and do not include percent intermediate (%I) in the statistic.

• Streptococcus pneumoniae and cefotaxime/ceftriaxone/penicillin: list %S using both meningitis and

nonmeningitis breakpoints; for penicillin, also indicate %S using oral breakpoint.

• Viridans group streptococci and penicillin: list both %I and %S.

• Staphylococcus aureus: list %S for all and methicillin-resistant Staphylococcus aureus (MRSA) subset.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 2

 Recommendations for Preparation, Education, and Solutions

Ideas on Educating Prescribers on Reading and Interpreting the Cumulative

Antibiogram
Do’s:
• Insert the antibiogram into the physician order entry computer program with links from the antibiotic
ordering screens

• Provide antibiograms on rounds with prescribers, fellows, residents, and students

• Implement antibiotic recommendations based on the antibiogram

• Facilitate development of antibiogram-related projects

• Develop a survey or quiz to assess antibiogram-related knowledge amongst clinicians

• Co-present the antibiogram at Medical Grand Rounds, P&T Committees, and other institutional meetings

Don’ts:
• Mail copies of antibiograms to prescribers as the only mechanism of dissemination and education

• Educate clinicians on the antibiogram only once each year when a new edition is ready

• Forget the opportunity to use additional tools to educate clinicians throughout the year on appropriate
empiric antibiotic therapy, such as newsletters, surveys, physician newsletters, and P&T agenda items

• Hide from clinicians the contact information of the ID Pharmacist, Microbiology, or Infection Prevention
when there are questions regarding the antibiogram

3 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Recommendations for Preparation, Education, and Solutions

Antibiogram Pitfalls and How to Fix Them

There are many pitfalls to antibiograms. These will result in confusion and potentially misinterpretation. A well-
defined antibiogram, while providing much more information, can also be cumbersome for clinicians. Therefore,
it is recommended that the following situations should be selected for improvement projects during the course of
antibiogram-related development, and not all will apply to your institution. However, each example below can
provide valuable information.
• Avoid testing antimicrobials using a cascade algorithm, such as testing restricted antimicrobial agents
only when resistant to first-line agents. Cascade testing (and subsequent reporting) is not the same as
selective reporting which is commonly employed in antimicrobial stewardship programs. See page 12.

• Use first isolate per patient in reporting period and include the method for eliminating duplicate isolates
(manually, or by altering an automated default exclusivity date, such as 7-day or 1-month or 1-year).

• Try to separate isolates of a bacterial species by differentiating patient location, source, and age. For
example:

o Adult vs. pediatrics

o Inpatient vs. outpatient/ED vs. long-term care

o Bacteremic isolates as a subset of overall results

o Adult ICU vs non-ICU vs. Hem-Onc service

o Urinary vs. non-urinary

o Isolates from patients with cystic fibrosis

• Develop a subset analysis for combination drugs against select species, such as Pseudomonas aeruginosa
and Acinetobacter baumanii

• Exclude from the cumulative antibiogram results obtained during surveillance studies (e.g.,
nasopharyngeal colonization studies for MRSA, VRE obtained from rectal swabbing, etc).

• Since antibiograms provide data for a single reporting period, it may be helpful to provide trending of
resistance (e.g., % of S,aureus isolates which are MRSA), %S results for medically important pathogens
in the institution (e.g., trends of ceftazidime resistance in P.aeruginosa), and the appearance of new
pathogens (e.g., KPCs as %K.pneumoniae or number of isolates overall). Pathogens not normally
included in the antibiogram but are medically important, such as Clostridium difficile, should be
communicated as part of an antibiogram in a separate section for educational value (see Template
examples).

• Regional or national cumulative antibiogram reports should carefully consider the demographics of
hospitals within the study, representation of data in quartiles of the mean, pooled median, ranges, and/
or the ability to compare institutions through risk stratification. Automated susceptibility testing methods
should be noted if differences exist.

• Institutional antibiograms do not provide information on how antibiotic use is epidemiologically linked to
resistance rates.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 4

 Scenarios and Examples

Contributions to Antibiotic Resistance May Be Out of Your Control:

The Importance of Patient Location on Antibiotic Susceptibilities
In the antibiogram pictured below, an institution (Hospital A) shows a % susceptible value of 79% for E.coli
(n=800 isolates) to Drug B. However, various sources contribute to this overall value and the number of first
isolates tested. The number of isolates contributed by inpatients at Hospital A consists of pediatric patients, adult
inpatients (non-ICU), and adult inpatients (ICU). When the sources of inpatient isolates (n=575) are considered,
the overall %S to Drug B is 92%, which contrasts sharply from the overall antibiogram results of 79% S. So
where is the additional resistance coming from?

During a pilot project, it is noted that E. coli isolates in patients from 3 local long-term care facilities exhibited
high resistance rates to Drug B. This came up during ICU rounds where 3 patients from LTCF B had been
admitted for urosepsis and each grew out E. coli from the blood and urine resistant to Drub B. All three patients
had been started by the ICU fellow on Drug B plus a single dose aminoglycoside (not Drug B).

The Antibiotic Stewardship Team (AST) approached the Microbiology laboratory to retrieve all test results from
the current antibiogram year for patients admitted from these LTC facilities and who showed positive cultures
for E. coli. The laboratory confirmed that all 225 isolates had contributed to the antibiogram. As a matter of
expediency, the Pharmacist selected 50 patient isolates at random. The antibiotic susceptibilities were calculated
and extrapolated according to the left-hand boxes above for LTCF B, C, and D.

It became immediately apparent that the high resistance rate of E. coli to Drug B was largely determined by
patient isolates from long-term care facilities but not from other inpatients within the hospital except for the adult
ICU (62% S). LTCF B appeared to be the “worst offender”.

5 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

As a quality improvement project, the AST approached the Medical Directors of all 3 LTCFs and asked if they
could assist the hospital in determining why resistance to Drug B was high. The Medical Directors and the
hospital AST assisted in developing appropriate recommendations for use of Drug B for the attending physicians,
infection prevention, and nursing at all 3 LTC facilities. The AST tracked E.coli susceptibillties to Drug B and 3
other agents over the following year.

A word of caution: since many patients in long-term care transition back-and-forth between hospital and nursing
home it may be difficult to determine the precise moment or location of acquisition of resistant pathogens. Not all
resistance is “imported”, but can be “exported” as well. A study such as that above should note this caveat.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 6

 Scenarios and Examples

The Problem with Antibiograms: Numbers Represent Single-Drug

Resistance
Numbers used in antibiograms represent susceptibilities, defined according to CLSI, for “bug-drug” combinations.
However, there may be clinical situations which in which drug combinations are necessary. These usually involve
the spectrum of drugs for which none are single drugs of choice (defined as any bug-drug combination with %S >
90%).

A common pathogen in which this frequently occurs is hospital-acquired infection with Pseudomonas aeruginosa,
especially bloodstream infection, pneumonia, and skin and soft tissue infection. In the antibiogram below, only
the susceptibility of amikacin exceeds 90%. While aminoglycosides are not thought of as monotherapeutic agents
for serious infections, the beta-lactams and ciprofloxacin would generally not be thought of as reliably active as
single agents. In such cases, the literature suggests that combination therapy may provide the best chances that
one of the two agents might be susceptible if the patient’s isolate in question is typical of the pseudomonads at this
institution.

One calculation and valuable piece of information for clinicians would be the construction of a cross-
susceptibility table. It is generally agreed that beta-lactams in combination with tobramycin or amikacin or
ciprofloxacin may satisfy the condition that one agent would demonstrate susceptibility. Arguments regarding
penetration of drugs into the pulmonary tissue is beyond the scope of this report.

% Susceptible
(2012 Antibiogram, respiratory tract, ICU adults)
Pathogen # Isolates
Piperacillin-
Ciprofloxacin Amikacin Ceftazidime
tazobactam
Pseudomonas
aeruginosa 100 84 70 92 78

In the table above, the susceptibilities of 100 isolates of P.aeruginosa are provided to four commonly used agents.
This data is also reflected in the cross-susceptibility table below. Cross-susceptibility tables are not included
in antibiograms, but the data can be valuable. To construct such a table isolates which are susceptible to both
agents must be determined (manually or with a program). The piperacillin-tazobactam/amikacin combination
provides a higher chance that the P.aeruginosa isolate is susceptible to both agents (S-S, 76%) compared to the
piperacillin-tazobactam/ciprofloxacin combination (S-S, 60%). Another view of this table is to determine the
chance that AT LEAST one agent of the 2-drug combination will demonstrate susceptibility. This is calculated by
adding the values for S-S, R-S, and S-R. For example, 100% of the isolates would be predicted to be susceptible
to either piperacillin-tazobactam or amikacin if these agents are combined. This is obvious because R-R for the
combination is zero (R-R = 0%). The chances that at least one of the agents of the beta-lactam plus ciprofloxacin
combination is susceptible to a group of 100 isolates is 94%.

7 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

Piperacillin-tazobactam S/R
(in combination with either Drug #1 or Drug #2)
Drug #1 S R Drug #2 S R

Ciprofloxacin
S 76 16 S 60 10
Amikacin

R 8 0 R 24 6

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 8

 Scenarios and Examples

Detecting Excessive Influence of Repeat (Duplicate) Isolates

The cumulative antibiogram should reflect a single isolate from one patient over the period of the report.
Generally, this isolate represents the first isolate collected from that patient in the data reporting period (e.g., one
year), as long as it is the same bacterial species irrespective of body site or susceptibility profile (phenotype).
This allows the antibiogram to be applied as a guide for selecting empiric antibiotic therapy. However, previous
susceptibilities from past admissions should be considered.

Susceptibilities may be biased if more than one isolate is collected from a patient. Culturing practices become
important in this case. Some clinicians may empirically select an antimicrobial agent against a likely pathogen
without culturing the patient, such as with UTIs in healthy young females. Unfortunately, in the institutionalized
patient, repeated culturing of the same site is common practice. Also, patients are more likely to have a therapeutic
failure related to either the patient’s inability to clear the infection (despite having selected an active antimicrobial
agent) or bacterial resistance. Therefore, a cumulative antibiogram report with many repeat isolates will generally
bias the results towards greater % resistance.

Note that there may be instances in which the resistance phenotype may differ; for example, E.coli #1 from blood
has a different phenotypic resistance pattern than E.coli #2 from urine. These may be counted as a single E.coli
isolate as long as the more resistant strain is counted (the antibiotic chosen should be directed at the bacteremic
isolate, and even if the urinary isolate is more resistant the drug concentrations achieved may overcome the MICs
of this strain).

One mechanism to calculate the potential influence of repeat isolates is to divide the # isolates of a particular
species by the number of unique patients during the antibiogram reporting period. Obviously, the ratio should be
1.00. However, it is not uncommon to find ratios of 2 to 3.

As the ratio increases there is a more likely chance that the cumulative antibiogram report will have greater %
resistance than first isolates only would. If the ratio is high work with the microbiology laboratory to eliminate
repeat isolates from the calculation prior to publishing the antibiogram. Also, the software program (Vitek,
Microscan, etc) can be reset to a longer period for eliminating duplicate isolates, such as 3 months or longer. It
might be interesting to compare susceptibilities and #isolates-to-patient ratio.

9 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

An example is provided below to demonstrate the potential influence of duplicate isolates on the %R of an
antibiogram.

SCENARIO: You are examining the %R of E.coli from urinary sources in both inpatients and patients
living in an attached spinal cord injury (SCI) unit. Ceftriaxone resistance is studied.
%R,
Ratio: %R, ceftriaxone
Institutionalized # Urinary E.coli # Unique
Isolate-to- ceftriaxone (elimination
Population Isolates Patients
Patient (raw data) of duplicate
isolates)
Hospital (400
2,000 1,500 1.3 4% 3%
adult beds)
SCI unit (100
1,000 200 5.0 27% 9%
adult beds)
EXPLANATION: Ceftriaxone may be an option more frequently, even in the SCI, once the isolate-
to-patient ratio is corrected. This may allow sparing of carbapenems in patients not felt to be
bacteremic. ESBLs may be more prevalent in populations with chronic indwelling Foley catheters
and frequent antibiotic exposures.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 10

 Scenarios and Examples

Antimicrobial Susceptibility Testing Using Cascade Algorithms: A Pitfall

Cascade testing refers to the antimicrobial susceptibility testing of an isolate only when it is resistant to a first-line
agent. This practice results in a bias in the resistance pattern to second-line and third-line drugs. Results should
be provided in the cumulative antibiogram report for isolates which are tested against all the agents in a specific
panel. Cascade reporting may be misinterpreted because such testing refers only to a subset of already-resistant
isolates and not to the entire isolate population collected during the study period.

Cascade testing and reporting is different from selective reporting, a commonly used strategy to report only
selected agents while suppressing the results of restricted antimicrobials.

In the example below, susceptibility results are provided for 4 drugs tested against Streptococcus pneumoniae.
However, the levofloxacin susceptibility result is determined (and reported) only for isolates resistant to
ceftriaxone. However, if susceptibility results are provided for all isolates of S.pneumoniae against all 4 drugs
the results are notably different. The second table provides a solution for reporting susceptibilities against key
pathogens.

% Susceptible
Pathogen # Isolates
Penicillin Azithromycin Ceftriaxone Levofloxacin
Streptococcus
pneumonia 100 87 63 91 90
(respiratory)

A 10% non-susceptibility rate for levofloxacin would be very unexpected in the U.S. and should alert clinicians to
an error in reporting. However, the 10% non-susceptibility rate (above) is derived from a single isolate tested only
when 10 MDR-SP isolates are tested against levofloxacin. Therefore, the true rate for susceptibility, ensuring that
no other isolates were levofloxacin-NS unless they were MDR, is 99% (below).

% Susceptible
Pathogen # Isolates
Penicillin Azithromycin Ceftriaxone Levofloxacin
Streptococcus
pneumonia 100 87 63 90 99
(respiratory)
# Isolates which were MDR and Resistant to Levofloxacin
MDR-SP* 10 -- -- -- 1
*MDR = non-susceptible to all three agents: penicillin, azithromycin, and ceftriaxone

11 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

When Antibiogram Data Fails to Provide Direction to Narrowing the

Antibiotic Spectrum in Select Patient Circumstances
While empiric antibiotic selection relies on the capability to predict the optimal antimicrobial regimen in a
specific individual patient, use of broad-spectrum agents will be common based on susceptibilities which provide
at least 90% susceptibility for suspected pathogens. It may appear that the antibiogram actually promotes the
selection of certain broad-spectrum agents such as piperacillin-tazobactam and carbapenems solely due to the fact
that their susceptibilities for many pathogens may exceed 90%.

A brief study by Gaynes et al. (see figure below) demonstrated that in the majority of situations it might not
be possible to target antimicrobial therapy based on culture results. Even in 135 cases in which initial empiric
piperacillin-tazobactam therapy was judged appropriate therapy could not be altered in the majority of patients.
The inability or failure to de-escalate after 72 hours of use arose from “indeterminate culture” results in 65% of
cases. Of these indeterminate cultures, 56% represented cultures obtained without growth or ‘normal flora’, 11%
of patients had no cultures obtained, and 33% of cultures were obtained after antibiotics were given. Suggestions
are provided in this study which can be employed selectively to narrow the spectrum of therapy.

Solutions may provide educational opportunities in the emergency department and admitting clinicians:
• Obtain cultures prior to antibiotic administration. When possible, note on the microbiology lab slip
any antibiotics given prior to admission and their timing, both oral and intravenous. In the ED, recent
antibiotic administration generally applies to intravenous agents given as soon as IV access is achieved.
Unfortunately, communication between healthcare workers does not consistently avoid this error.
Patients who cannot produce a high-quality sputum sample should be induced as soon as possible using
respiratory therapists and protocols for sputum induction prior to administration of antibiotics. While
antibiotic distribution into tissues is time-dependent even orally administered antibiotics can be absorbed
fairly rapidly in some individuals.

• The lack of culture results within 24 hours of admission may largely represent cultures “lost” or not
processed. Everyone has seen cultures in the ED which were not sent to the laboratory, not placed on ice,
or not otherwise drawn into the appropriate tubes. A process and time-labor analysis may provide ideas
for process improvement.

• Blood culture contamination is not infrequent albeit low (2-4% is commonly cited, but rates may be
significantly higher). Along with the practice of ‘swabbing non-sterile wounds or tissue’ there is little
wonder why coagulase-negative staphylococci represent large numbers of isolates on many antibiograms.
Appropriate site cleansing prior to drawing blood cultures has reduced contamination rates significantly
and should be studied as a process improvement project. Blood draws from an existing IV line may also
increase blood contamination rates. Again, wounds cultures should be taken from deep tissues and not just
‘swabbed’.

• When bacteremia due to a vascular source is suspected, submit blood cultures with instructions to plate
on nutritionally-supplemented media (viridians streptococci) and hold such cultures for a longer period of
time (according to the laboratory’s protocol). Rapid identification of slow-growing gram-positive or gram-
negative pathogens can reduce hospitalization days and possibly complication from persistent bacteremia
which has not been documented by culture results.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 12

 Scenarios and Examples

13 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

Presenting Multi-Institutional Cumulative Antibiogram Data:

Local, Regional, and National Results
While an antibiogram from a single institution is one of the key activities of an antibiotic stewardship program,
there may be opportunities to construct an antibiogram representing several or even hundreds of institutions.

There have been many difficulties associated with compiling such antibiograms, including quality assurance and
data verification of results prior to calculating average percent susceptibilities. The most difficult challenge has
been to risk stratify reporting institutions because multi-institutional antibiograms may represent a large array
of hospitals and long-term care facilities. For example, antibiogram data from a large academic hospital which
performs organ and hematopoietic transplants might be expected to have more resistance compared to a small
community non-teaching hospital with less association with long-term care facilities.

While many examples are provided in the literature, a multi-institutional antibiogram should contain the following
elements, and efforts should ensure that certain data can be acquired from all participants:
• Laboratory testing methodology

• MIC breakpoints used to interpret S, I, and R.

• Representation of the key pathogens, such as E.coli, K.pneumoniae, Enterobacter spp., P.aeruginosa, S.
aureus (including MRSA), S.pneumoniae, and Enterococcus spp.

• Risk score for each institution, such as case-mix index; or, demographic categorization of each hospital

• Percentiles at 10%, 25%, 50% (median), 75%, and 90% are useful to identify outliers. In the table
below, percentiles are most useful for large numbers of institutions, such as state and national data.
This methodology is used by the CDC’s NHSN program. For example, at the 25th percentile, 25% of
the hospitals had lower %S rates and 75% of the hospitals had higher %S rates. If the %S rate is below
the 25th percentile, determine whether it is below the 10th percentile. If the %S rate is, then it is a low
outlier which may be due to resistance issues within the institution or feeding of high numbers of patients
from SNFs where resistance may be a greater problem. Regardless, the institutions with more resistance
deserve attention. Of course, there may also be reporting inaccuracies and these should be ruled out first.
On the other side of this is the institution included in the 90 percentile for %S (i.e., 90% of the hospitals
in the data set have %S rates which are lower than the 90 percentile performers. For top performing
institutions, these should be analyzed for efficiency of the antibiotic stewardship program. These best
practice centers can be targeted for duplicating antibiotic prescribing which might reduce resistance in
other institutions.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 14

 Scenarios and Examples

Percentile† (%S ceftriaxone;

%S, ceftriaxone
# # Institutions K.pneumoniae)
Pathogen
Isolates Reporting Pooled Range 50%
10% 25% 75% 90%
mean (hi. lo) (median)
K.pneumonaie
All 2,888 7 85 61, 94 na 75 90 95 na
Urinary 2,510 7 84 58, 94 na 74 89 94 na
Non-urinary 378 7 93 79, 100 na 81 96 99 na
† N=7. Due to the low number of sites, percentiles were interpreted as follows. For all isolates, 1 institution (~10th percentile) had %S
of 61%, 2 institutions (~25th percentile) had susceptibility rates of 75% or lower, and 5 institutions (~75th percentile) had susceptibility
rates of 92% or lower. The 50% value (median) is the middle value of 7 institutions; therefore, for all isolates considered amongst 7 sites,
3 institutions had % susceptibility rates which were lower than 90% (all isolates), 89% (urinary isolates), and 96% (non-urinary), and 3
institutions had %susceptibility rates which were higher than these median values.

15 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

Presenting Trends in Resistance As An Educational Section of the

Antibiogram
A challenge with antibiograms is the presentation of data which is restricted to a specific one-year period.
This creates a lack of perspective, such as trends in resistance rates or the rise/decline of pathogens which are
medically important.

An appropriate perspective reflecting such trends can be reserved for a section of the antibiogram which is not
commonly used for educational purposes – the margin.

• Due to space limitations, the number of graphs and tables should be minimized.

• Alternatively, these may be presented as part of an extended and ongoing educational program which aim
is to provide perspective in resistance through trending over 5-year (or greater) periods.

• Also, the rise of medically important pathogens which have not been observed in past years is very
valuable for both close observation and as a patient safety issue as frequently such organisms present few
options for selection of antimicrobials.

• Pathogens of epidemiologic importance should also be presented, such as Clostridium difficile and
isoniazid-R and rifampin-R strains of Mycrobacterium tuberculosis.

Examples are provided below:

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 16

 Scenarios and Examples

Institutional Antibiograms Do Not Provide Information On How Antibiotic

Use is Epidemiologically Linked to Resistance Rates
Cumulative antibiogram reports do not provide details as to the density of antibiotic use and therefore adverse
trends in resistance or the emergence of new MDR pathogens cannot be directly linked to the local use of a
specific antibiotic or class. Similarly, antibiograms may not provide evidence that local microbiologic outcomes
are a result of changes in antibiotic use patterns, although it is the implied function of the antibiotic stewardship
program. While antibiotic use applies selective pressure for the emergence of resistance, there is little data to
guide clinicians of an ASP to determine how resistance trends can be altered. There are several limitations to
studies which highlight the complexity between antibiotic use and bacterial resistance even when both are studied
as local and simultaneous occurrences.
• Resistance and high antibiotic density may not occur in the same hospital unit

• Changes must be studied over a period of several years; a change in resistance in the recent antibiogram
should not be assumed to result from changes in antibiotic pressures. More sophisticated methods should
be employed, such as interrupted time-series analyses.

• Outside influences, such as imported resistance, is not accounted for in an antibiogram.

• Antibiograms do not assess the changes in MICs since S, I, an R are determined by breakpoints.
Migration of MICs towards the breakpoint is a harbinger towards resistance.

• Hospital-wide antibiograms do not provide detailed analysis of specialized areas of the hospital or
patient subpopulations. Therefore, emerging pockets of resistance may be missed. The overall bug-drug
susceptibility values dilutes out the effect of an emerging resistance problem.

• Antibiograms constructed using a first-isolate method underestimates true resistance since susceptible
isolates may be replaced by resistant ones during therapy.

• Antibiograms do not capture multidrug resistant organisms; only single bug-drug combinations are
represented.

• Resistance may be curtailed through other measures, such as infection prevention measures, and includes
improvements in hand hygiene, patient isolation and patient movement protocols, improved technology
for insertion and maintenance of entry sites for central lines, and updated room cleaning procedures.

• Changes in antibiotic policies frequently result in “squeezing the balloon” whereby restriction of one
antibiotic class results in over-utilization of another.

The deficiencies of an antibiogram based on its static nature of analysis should not discourage clinicians involved
with ASP activities from creating a more innovative and useful tool.

17 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

Assessing Resistance Trends:

Utilizing Statistical Analysis to Evaluate Changes in Susceptibility Rates
Historic annual antibiogram data can be an invaluable tool to help track trends in changing resistance and
indicate a need for further investigation and potential action. Additionally, comparisons can be made from within
institutions (from inpatients to outpatients) or externally (from one institution to regional or national data).
Generally, this is determined by evaluating changes in %S estimates between different data sets for specific
organisms and antimicrobials. A crucial part of analysis is determining the precision of a %S estimate and the
significance of an increase or decrease in susceptibility in order to identify the need for action.

A confidence interval is used to provide an estimate of how precise the observed %S is when used to guide
clinical decision making. The sample size (number of isolates tested) influences the precision of the estimate
and the subsequent confidence interval. The larger the sample size, the more precise the resulting observed %S;
the smaller the sample size, the less precise. This serves to validate the %S value and allows the data analyst to
determine with what confidence the observed %S represents the broader population.

One common statistical test utilized to determine statistically significant differences in resistance rates is the Chi-
squared test. Generally, a P value of < 0.05 is accepted to indicate that the observed differences are not likely due
to chance. Information about Chi-squared calculations can be found in biostatistics textbooks, however, the CLSI
M39-A3 consensus document has appendices that may be used as a guide to determine statistical significance.
Keep in mind, the tables provided in this document can only be used if the two populations being compared are of
similar sample size.

While analysis of resistance trends as described above can identify “statistically significant” differences, this
should not be confused with or imply a “clinically or epidemiologically important” difference. In the case of a
large number of isolates (sample size), small changes in %S such as a decrease from 63.2% to 61.9% may be
statistically significant, but deemed unimportant when evaluating in regards to clinical application. Conversely,
in the case of a small number of isolates, a change in %S from 80% to 55% may not be statistically significant
but clinically may alert the institution to
a potential emergence of resistance. In
both of these cases, the institution(s)
must determine whether the results are
due to true changes in susceptibility or
confounded by other factors including
changes in the patient population, sample
collection practices, laboratory testing or
data reporting.

Regardless of the method used, critical

analysis of changes in antimicrobial
resistance patterns using antibiogram data
can help identify areas of improvement
related to antimicrobial prescribing and
provide a focus for stewardship activities.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 18

 Scenarios and Examples

Using the Antibiogram as Part of Antimicrobial Stewardship Initiatives

Antibiograms have a variety of applications to clinical practice and data gathered can help identify potential
opportunities for improved antimicrobial prescribing. While practitioner education plays a key role in prescribing
practices, improvement can also be realized through targeted initiatives. This goal can be achieved through
several different stewardship efforts which vary in complexity when considering implementation and impact.
The following are some select examples of how antibiogram data can be incorporated into stewardship-related
activities.

Formulary Considerations:
• In response to increasing resistance trends, institutions may consider formulary changes using
antibiogram data as a guide. Often these involve changing agents within the same class.

• A study by Empey et al described a significant decrease in the observed rates of ceftazidime-resistant

Pseudomonas aeruginosa, ceftazidime-resistant Klebsiella pneumoniae and piperacillin-resistant
Pseudomonas aeruginosa infections in patients after changing their cephalosporin formulary from
ceftazidime and cefotaxime to cefepime.

Antibiotic restriction:
• Based on antibiogram susceptibility trends, use of specific agents or classes of agents may be restricted or
controlled. Traditionally this has applied to broad-spectrum agents but could be individualized based on
local antibiogram data. Prescribers must obtain prior approval in order to use the restricted agent.

Prospective review:

• Similar to antibiotic restriction, this intervention identifies targeted agents based on resistance trends and
aims to decrease use. However, the method employed here utilizes a back end approach which requires an
infectious diseases expert to review all uses of the prescribed agent and make recommendations in order
to decrease inappropriate use and impact resistance rates.

Order Set/Clinical Pathway Design:

• Antibiogram data and trends can be incorporated into the design of hospital-specific order sets,
guidelines, and clinical pathways in order to increase or decrease empiric use of specific agents based on
susceptibility.

• An example of this would be developing empiric antibiotic selections as part of a severe sepsis admission
order set. Based on the hospital antibiogram, cefepime, piperacilllin/tazobactam, and tobramycin have
consistently high susceptibilities to most gram-negative organisms, including Pseudomonas aeruginosa.
Comparatively, fluoroquinolones may demonstrate lower gram-negative susceptibilities and these
susceptibilities have continually trended downward over the past several years. Using this information,
the order set could be built to include only cefepime and piperacillin/tazobactam as primary first line
agents for gram-negatives. While fluoroquinolones may be excluded from the selection list, tobramycin
can be included as an adjunct agent. Again, empiric first-choice antibiotics will be based upon the
antibiogram, but a more focused examination of previous cultures obtained from septic patients may be
warranted since this is the target population.

19 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Scenarios and Examples

Computer-assisted decision support services (CaDSS):

• Some institutions have the ability to embed predefined pathways and restrictions on antimicrobial
selection electronically as part of the ordering process.

• A study by Pestotnik et al found that a computer-assisted decision support program resulted in an overall
reduction antibiotic use, of 22.8% over the study period. The institution’s antibiogram remained stable
over the 7-year period.

It should be noted that often these stewardship initiatives are established with the two-pronged goal of improving
patient outcomes and improving resistance rates. While specific patient outcomes can be measured, it is more
difficult to assess the true impact of a specific stewardship initiative on changes in rates of resistance, as these
often appear months or years after an intervention and can be influenced by a number of factors. Additionally,
decreasing the use of one or more antibiotics will invariably cause an increase in use of another agent or class of
agents. It is important to take the susceptibility changes for these other agents into account when assessing impact.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 20

 Reading List

A Brief Reference List on Antibiograms

Binkley S, Fishman N, LaRosa L, et al. Comparison of unit-specific and hospital-wide antibiograms: potential
implications for selection of empirical antimicrobial therapy. Infect Control Hosp Epidemiol. 2006;27(7):682-7.

Christoff J, Tolentino J, Mawdsley E, et al. Optimizing empirical antimicrobial therapy for infection due to
gram-negative pathogens in the intensive care unit: utility of a combination antibiogram. Infect Control Hosp
Epidemiol. 2010;31(3):256-61.

CLSI. Analysis and Presentation of Cumulative Antimicrobial Susceptibility Test Data; Approved Guideline-
Third Edition. CLSI document M39-A3. Wayne, PA: Clinical Laboratory Standards Institute; 2009.

D’Agata E, Cataldo M, Cauda R, Tacconelli E. The importance of addressing multidrug resistance and not
assuming single-drug resistance in case-control studies. Infect Control Hosp Epidemiol. 2006;27(7):670-4.

Dellit TH, et al. Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America
Guidelines for Developing an Institutional Program to Enhance Antimicrobial Stewardship Clin Infect Dis.
2007;44:159-77

Empey K, Rapp R, Evans M. The effect of an antimicrobial formulary change on hospital resistance patterns.
Pharmacotherapy. 2002;22(1):81-7.

Gaynes R, Gould C, Edwards J, et al. A multicenter study on optimizing piperacillin-tazobactam use: lessons on
why interventions fail. Infect Control Hosp Epidemiol. 2009;30(8):794-6.

Hindler J, Stelling J. Analysis and presentation of cumulative antibiograms: a new consensus guideline from the
Clinical Laboratory Standards Institute. Clin Infect Dis. 2007;44:867-73.

Kuper KM, et al. Antimicrobial susceptibility testing: a primer for clinicians. Pharmacotherapy 2009;29:1326–
1343.

McGregor J, Bearden D, Townes J, et al. Comparison of antibiograms developed for inpatients and primary care
outpatients. Diagn Microbiol Infect Dis. 2013;76:73-9.

Pakyz A. The utility of hospital antibiograms as tools for guiding empiric therapy and tracking resistance.
Pharmacotherapy. 2007;27(9):1306-12.

Pestotnik SL, et al. Implementing antibiotic practice guidelines through computer-assisted decision support:
clinical and financial outcomes. Ann Intern Med 1996;124:884-890.

Schulz L, Fox B, Polk R. Can the antibiogram be used to assess microbiologic outcomes after antimicrobial
stewardship interventions? A critical review of the literature. Pharmacotherapy. 2012;32(8):668-76.

Slain D, Sarwari A, Petros K, et al. Impact of a multimodal antimicrobial stewardship program on Pseudomonas
aeruginosa susceptibility and antimicrobial use in the intensive care unit setting. Crit Care Res Pract. 2011;
Article ID 416426:1-5; doi:10.1155/2011/416426.

21 Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit

 Appendix

THE CUMULATIVE ANTIBIOGRAM REPORT: TEMPLATES & SUGGESTIONS

The following templates for antibiograms serve as
examples for reporting antimicrobial susceptibilities
against gram-negative and gram-positive pathogens
St. Elsewhere Medical Center
(isolates ≥30 per reporting period) with a separate Antimicrobial Susceptibility
template for Streptococcus pneumoniae. Summary
2012 Calendar Year
These templates are best adapted to an 8” x 14” legal
sized sheet which can be doubly folded into a convenient Antimicrobial Stewardship Program
3.5” x 8” form. Clinical Microbiology; Dept of
Pathology & Laboratory Medicine
A face page listing contact information of key clinicians
should be listed (see panel to right). Phone or pager The information contained in this
numbers of key contact personnel should be provided summary can also be found on the
as well as those of the Antimicrobial Stewardship “Antibiogram” link on the hospital
Committee and ancillary personnel such as the ID service intranet:
for consultation, Infection Prevention, Microbiology, ID http://www.stelsewhereASP.edu/
Pharmacist, Drug Information, and Central Pharmacy.
For questions concerning antibiotic
Although drug costs have been traditionally included susceptibilities or antibiogram
on antibiograms (i.e., a plus-sign scoring system or interpretation, please call:
wholesale acquisition cost for standard dose sizes) the Edward E. Coli, Clinical Microbiology
conversion of branded products to generics, multiple Supervisor:
manufacturers of generics, drug shortages, and pharmacy
555-555-5555
contract pricing and rebates makes assignment of prices
Sally M. Onella, PharmD., ASP
even by a ranking system obsolete. A focus on accurate
Pharmacist:
empiric drug therapy directed through antimicrobial
555-555-5556 (ID pager)
susceptibilities and interpretations should be the primary
focus of the antibiogram. Education of clinicians
regarding antimicrobial prescribing practices rather than
pharmacy expenditures makes the antibiogram more
Frequently called numbers:
valuable. Empiric antimicrobial recommendations can
Antimicrobial Testing Laboratory
be included for major infections, such as pneumonia, 555-555-5557
cellulitis, COPD exacerbation, urinary tract infection, Infectious Diseases (Adult)
and sepsis/septic shock. These may be further divided 555-555-5558
into community-associated and hospital-associated Infectious Diseases (Peds)
infections, such as hospital-acquired or ventilator- 555-555-5559
associated pneumonia (HAP or VAP, respectively), febrile Drug Information Center
neutropenia, and central-line infections with suspected 555-555-5560
bacteremia. Infection Prevention
555-555-5561
Hospitals with restrictive formularies may indicate Inpatient Pharmacy
antimicrobials which require ID approval or other 555-555-5562
approved formulary indications.

Arizona Healthcare-Associated Infections (HAI) program — Antibiogram Toolkit 22

 Key Notes on % Susceptible for Gram-Negative Isolates
St. Elsewhere
USE OF IN VITRO DATA MUST BE INTERPRETED IN REFERENCE TO CLINICAL STATUS, SITE OF INFECTION, AND
Antimicrobial Susceptibilities PENETRATION OF THE ANTIMICROBIAL ANTIBIOTIC
Medical Center
[This section should be included in
antibiograms to discuss important trends in
2012 Antibiogram
resistance for important pathogens, graphs
Isolates, Jan - Dec 2012
or tables which track antimicrobial use or
changes in MDRO patterns, Clostridium
difficile infection trends and information,
% Susceptible
KPC and MRSA incidences, and #

Amikacin
Gentamicin
Tobramycin
Ampicillin
Ampicillin-ulbactam
Piperacillin-
tazobactam
Aztreonam
Cefazolin
Ceftriaxone
Cefepime
Ciprofloxacin
Meropenem
Trimethoprim-
sulfamethoxazole
Nitrofurantoin

Organism
recommendations for interpreting Strains
susceptibilities with new agents or testing Escherichia coli
systems. New dosing recommendations may Klebsiella
be suggested, such as for colistin, pneumoniae
aminoglycosides, or vancomycin] Enterobacter
cloacae
Enterobacter
aerogenes
Proteus mirabilis
Salmonella spp

Miscellaneous Susceptibility Shigella spp

Information Citrobacter
freundii

[Testing for anaerobes and TB may be Acinetobacter

baumannii
intermittent but remain relevant in some
Pseudomonas
centers. Important reminders may be stated aeruginosa
in this section in efforts to extend Stenotrophomonas
antimicrobial stewardship principles, etc] maltophilia

[Appropriate footnotes may be included in this box with reference to either antimicrobials or pathogens. For example, it should be noted that
nitrofurantoin is prescribed for urinary tract infections only. Also, dosing of piperacillin-tazobactam for treatment of serious pseudomonal infections
should consider higher dosages and/or extended infusions, if applicable]
 Key Notes on Gram-Positive % Susceptible for Gram-Positive Isolates
Antimicrobial Susceptibilities USE OF IN VITRO DATA MUST BE INTERPRETED IN REFERENCE TO CLINICAL STATUS, SITE OF INFECTION, AND PENETRATION OF THE
St. Elsewhere ANTIMICROBIAL ANTIBIOTIC

[This section should be included in

Medical Center ALL Isolates (from all patients & sources) Blood Isolates (from all patients) *
antibiograms to discuss important
trends in resistance for important 2012 Antibiogram
pathogens, such as Staphylococcus
Isolates, Jan - Dec 2012
aureus and vancomycin-resistant
Enterococcus].
% Susceptible
[An expanded column for all strains
and blood isolates is included in this #
template, but may also be applied to Strains
Organism
the template for gram-negative (all/

Clindamycin
Doxycycline
Erythromycin
Oxacillin
Penicillin
Rifampin
Trimethoprimi-
Sulfamethoxazole
Vancomycin
Ampicillin
Daptomycin
Linezolid
Oxacillin
Vancomycin
Gentamicin (SYN)
Streptomycin (SYN)

pathogens. Also note this template blood)

also distinguishes location of patient Staphylococcus 1473/
-- --
at time of specimen collection. aureus (all) 107
While this is only an example for S. Outpatient 781 100 -- --
aureus, it may be considered for Inpatient 461 100 -- --
other pathogens which may be
ICU 231 100 -- --
subject to specific hospital-acquired
infection reporting]. Methicillin- 625/
resistant S. aureus 41
Coagulase-
[New dosing recommendations may 1005/
negative
§ 121
be suggested in this box, such as staphylococcus
vancomycin dosing to achieve higher Viridans group
37/
trough serum concentrations or low- Streptococcus -- -- -- -- -- -- --
(10)
dose aminoglycosides when (sterile sites only)
Enterococcus 425/
appropriate for treatment of 54 62
faecalis † 32
infective endocarditis]. Enterococcus 94/
‡ 61 60
faecium (12)
[Important reminders may be stated Examples of footnotes for gram-positive pathogens follow:
in this section in efforts to extend * Includes agents acceptable for treatment of bloodstream infection. ID consult is recommended for bloodstream infections due to gram-positive
antimicrobial stewardship infections.
principles]. † 19% high-level resistance to both GEN SYN and STR SYN ‡ 44% high-level resistance to both GEN SYN and STR SYN
§ Excludes S. lugdunensis and S. saprophyticus
( ) Less than 30 isolates; susceptibility results are not provided
 Key Notes on Streptococcus pneumonia % Susceptible for Streptococcus pneumoniae
St. Elsewhere USE OF IN VITRO DATA MUST BE INTERPRETED IN REFERENCE TO CLINICAL STATUS, SITE OF
Antimicrobial Susceptibilities INFECTION, AND PENETRATION OF THE ANTIMICROBIAL ANTIBIOTIC
Medical Center
[This section should be included in antibiograms to discuss
important trends in resistance for this pathogen]
2012 Antibiogram
Isolates, Jan - Dec 2012
[For few other pathogens are susceptibilities and their reporting so
confusing as with Streptococcus pneumonia. This is often related
% Susceptible
to differing breakpoints depending on whether the pathogen is
isolated from CSF as with bacterial meningitis or from blood as in #

Amoxicillin (PO)
Cefotaxime
Ceftriaxone
Clindamycin
Erythromycin
Levofloxacin
Moxifloxacin
Penicillin (IV)
Penicillin (PO)
Trimethoprim-
sulfamethoxazole
Vancomycin

Organism
pneumococcal pneumonia with bacteremia. Note that the Strains
number of strains is 100 but the same for non-meningitis and S. pneumoniae
100 --a --a --a 64§ 100
meningitis. Thus, this presentation expresses all isolates in terms (ALL)
† † †
of %susceptible applied to both breakpoints. It is important to Non-meninigitis 100 -- 94 95 -- -- -- -- 84 -- -- --
note in this box the number of pneumococcal isolates from CSF Meningitis 100 -- 85‡ 84‡ -- -- -- -- 64‡ -- -- --
and non-CSF specimens. Also, the antibiogram template may be Examples of footnotes for Streptococcus pneumonia include the following examples:
expanded to additional rows to include adult and pediatric data
a
separately] Breakpoints differ for cefotaxime, ceftriaxone, and penicillin based on diagnosis
†
Susceptible breakpoint for S.pneumoniae in patients with meningitis is ≤ 0.5 mg/L for cefotaxime and ceftriaxone and
[Dosing recommendations may be suggested in this box, especially ≤ 0.06 mg/L for penicillin
‡
Susceptible breakpoint for S.pneumoniae in patients with nonmeningitis infections is ≤ 1 mg/L for cefotaxime and
when dealing with the beta-lactam agents and vancomycin for the ceftriaxone and ≤ 2 mg/L for penicillin
treatment of meningitis] §
Susceptible breakpoint for S.pneumoniae is ≤ 0.06 mg/L for penicillin when penicillin V is administered by the oral route