1. No category

Educational Workshop
EW12: Know your own local antibiotic susceptibility data
arranged with the ISC Working Group on Antimicrobial
Resistance
Convenors:
Ian M. Gould ((Aberdeen,, UK))
Wing-Hong Seto (Hong Kong, HK)
Faculty:
Sylvain DeLisle (Baltimore,
(Baltimore US)
Niels Frimodt-Møller (Copenhagen, DK; no
data submitted)
Jose Maria Lopez Lozano (Alicante, ES)
Po-Ren Hsueh (Taiwan,
(
CN))
DeLisle – Local antibiotic susceptibility data - USA
Incorporating the Antibiogram into
Decision Support: a ProcessProcess-based
Approach
Sylvain DeLisle M.D.,M.B.A
Special Assistant to the COS, VA Maryland HCS
Assoc. Prof. of Medicine
University of Maryland, Baltimore, MD
[email protected]
Objectives
• Gain a conceptual framework for the
pharmacotherapeutic process
• Gain
G i an approach
h tto id
identify
tif h
how
antibiograms can be used to optimize
antibiotherapies
• Prioritize steps needed to automate both
construction and sharing of aggregated
antibiogram
Last week, in my ICU
• 50 yo male with ALS, on MV since 5/10 at
a chronic care facility
• 2 pastt episodes
i d off VAP
VAP, llastt one iin 3/11
• Presents with fever, worsening
oxygenation, MAP 60 despite 3L IVF,
WBC 14K with left shift, new LLL CXR
infiltrate
3
DeLisle – Local antibiotic susceptibility data - USA
What is unique about antibiotherapies
antibiotherapies?
?
• Dual responsibility to
– Treat this patient effectively
– Preserve the effectiveness of antibiotics
for future patients
How Can We Avoid Both Underuse
and Overuse of Antibiotics in this
Patient?
N
Stop
Drug
Needed?
Y
Y
Choose Drug
N
Safe/
Effective
Evaluate Drug
Deliver Drug
Prescribe Drug
4
DeLisle – Local antibiotic susceptibility data - USA
N
Stop
Drug
Needed?
Y
Y
Safe/
N
Effective
Choose Drug
Evaluate Drug
Deliver Drug
Prescribe Drug
Are Antibiotics Needed?
• Initially, infections are probabilistic diagnoses
• SIRS criteria
– Temp (<36, >38ºC)
– HR >90
– RR >20
– WBC (<2k, >12k)
• Localizing symptoms/signs
• Suggestive test/imaging results
Are Antibiotics Needed?
Non--infectious causes of Fever
Non
•
•
•
•
•
•
•
•
•
Acalculus cholecyst.
cholecyst.
Acute MI
Adrenal Insuff.
Insuff.
CVA
Dressler’s syndrome
Drug fever
DVT
Fat emboli
Gout
• Immune
reconstitution
• Intracranial Bleed
• Pancreatitis
• Pulm Infarct
• Pneumonitis
• Thyroid Storm
• Transfusions
5
DeLisle – Local antibiotic susceptibility data - USA
Mermel, Clin Infect Dis 49, 1 (2009)
Last Week, in my ICU
• Patient will need antibiotics…
Timely Drug Delivery
• Retrospective record review
• 2731 patients with septic shock 55-10 yr period
• 9 ICUs, 7 hospitals (2 tertiary, 5 community)
Kumar, Crit Care Med 354:1589 (2006)
6
DeLisle – Local antibiotic susceptibility data - USA
Timely Drug Delivery
Kumar, Crit Care Med 354:1589 (2006)
Last Week, in my ICU
• Patient will need antibiotics…
• …quickly
Last Week, in my ICU
• Sending blood, urine, tracheal aspirate,
mini--BAL cultures
mini
• Aggressive
A
i IV fluid
fl id resuscitation
it ti to
t CVP off
8, starting pressors if MAP remains below
< 65. Measuring lactate clearance.
7
DeLisle – Local antibiotic susceptibility data - USA
N
Stop
Drug
Needed?
Y
Y
Choose Drug
Safe/
N
Effective
Evaluate Drug
Deliver Drug
Prescribe Drug
N
Stop
Drug
Needed?
Y
Y
Choose Drug
N
Safe/
Effective
Evaluate Drug
Deliver Drug
Prescribe Drug
Choosing Appropriate Empiric
Antibiotics: Input Variables
• Environment
• Location(s
Location(s)) of care
• Country, Region, City, Healthcare, Hospital, Ward
• PatientPatient-level
• Co
Co--morbidities, specific risk factors, immune status
• Past and current source of infections
• MicroorganismMicroorganism-level
• Colonization status
• Past organisms and sensitivities
8
DeLisle – Local antibiotic susceptibility data - USA
Mortality Associated with Inadequate
Initial Therapy
Alvarez-lerma,…
Rello, 1997
Kollef, 1999
Kollef, 1998
Ibrahim, 2000
Luna, 1997
0
10
20
30
40
50
60
70
80
90
100
%Mortality
Initial Adequate herapy
Initial Inadequate Therapy
Risk Factors for Inadequate Empiric
Antibiotics:
• Organism:
– VRE (100%)
– Candida (95%)
– MRSA ((33%),
), CNS (22%),
(
), Pseudomonas (10%)
(
)
Ibrahim, Chest 118:146 (2000)
•
Severity of illness
– Narrower spectrum in less severe disease
N
Stop
Drug
Needed?
Y
Y
Choose Drug
N
Safe/
Effective
Evaluate Drug
Deliver Drug
Prescribe Drug
9
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Y
Choose Drug
Safe/
N
Effective
N
Non-Non
N On the
Formulary
Formulary
Req.
N
Justified?
Y
Y
Y
Restricted?
N
Y
N
Best
Practice?
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Y
Right
N
Frequency/
Dose
Prescribe Drug
Formulary Restrictions to Adjust
Empiric Antibiotics
Rahal, JAMA 280:1233 (1998)
Formulary Restrictions to Adjust
Empiric Antibiotics
Rahal, JAMA 280:1233 (1998)
10
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Y
Choose Drug
Safe/
N
Effective
N
Non-Non
N On the
Formulary
Formulary
Req.
N
Y
Evaluate Drug
Y
Y
Ri
Right
ht
Patient/
Time
Y
Restricted?
Justified?
Deliver Drug
N
Y
N
Y
N
Best
Practice?
Right
N
Frequency/
Dose
Y
Prescribe Drug
N
Drug
Needed?
Stop
Y
Y
Choose Drug
N
Safe/
Effective
N
NonNonN On the
Formulary
Formulary
Req.
N
Y
Justified?
Y
Y
Restricted?
N
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Y
N
Best
Practice?
Y
Right
N
Frequency/
Dose
Prescribe Drug
Choosing Appropriate Empiric
Antibiotics: AMT
Gross, Clin Infect Dis 33:289 (2001)
11
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Y
Choose Drug
Safe/
N
Effective
N
Non-Non
N On the
Formulary
Formulary
Req.
N
Y
Y
Y
Restricted?
Justified?
N
Y
N
Best
Practice?
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Y
Right
N
Frequency/
Dose
Prescribe Drug
Appropriate Empiric Antibiotics
Effect of Sepsis Bundle
Barochia, Crit Care Med 38:668 (2010)
Last Week, in my ICU
• Past antibiograms have revealed MRSA,
XTR Klebsiella
Klebsiella,, E. coli, Acinetobacter
• Started
St t d empirically
i i ll on vancomycin,
vancomycin
i ,
pip/tazo
pip/
tazo,, tigecycline,
tigecycline, colistin
• Seen by Infectious Disease
12
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Y
Choose Drug
Safe/
N
Effective
N
Non-Non
N On the
Formulary
Formulary
Req.
Y
N
Justified?
Y
Y
Restricted?
N
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Y
N
Best
Practice?
Y
Right
N
Frequency/
Dose
Prescribe Drug
Right Posology
Posology,, Right Route
• Pharmacokinetic and pharmacodynamic principles,
dosage adjustments for sensitivies
sensitivies,, renal/liver function,
considerations for drugdrug-drug interactions, should all be
incorporated in guideline development
• More efficient conversion from IV to oral therapy is often
possible, saves money and thus may help fund
antimicrobial management program
Converting from IV to PO
• 541 patients admitted for pneumonia
• Oral co
co--amoxiclav vs. IVIV-thenthen-oral co
co--amoxiclav vs. IVIVthenthen-oral cefuroxime
• LOS reduced by 2 days
Chan, BMJ 310:1360 (1995)
13
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Y
Choose Drug
Safe/
N
Effective
N
Non-Non
N On the
Formulary
Formulary
Req.
N
Y
Evaluate Drug
Y
Y
Ri
Right
ht
Patient/
Time
Y
Restricted?
Justified?
Deliver Drug
N
Y
N
Y
N
Best
Practice?
Right
N
Frequency/
Dose
Y
Prescribe Drug
Timely Drug Delivery
Effect of Sepsis Bundle
Barochia, Crit Care Med 2010, 38, 668
N
Drug
Needed?
Stop
Y
Y
Choose Drug
N
Safe/
Effective
N
NonNonN On the
Formulary
Formulary
Req.
N
Justified?
Y
Y
Y
Restricted?
N
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Y
N
Best
Practice?
Y
Right
N
Frequency/
Dose
Prescribe Drug
14
DeLisle – Local antibiotic susceptibility data - USA
Distinguishing Infection from
Colonization: Rapid Diagnostic Testing
• PNA FISH assay allows identification of S. aureus vs.
CoNS in blood cultures in ~3 hours.
• AMT evaluated PNA FISH group, but not the control
group
Forrest, G Antimicrob Chemother 58:154 (2006), slide courtesy of S. Cosgrove
Revising Antibiotics Regimen
De--Escalation in VAP
De
Rello, Crit Care Med 32:2183 (2004)
Revising Antibiotics Regimen
Quantitative Cultures
Fagon, Ann Intern Med 132:631 (2000)
15
DeLisle – Local antibiotic susceptibility data - USA
Revising Antibiotics Regimen
Clinical Pulmonary Infection Score
• CPIS criteria:
– Temperature (>38.5, >38.9 or <36)
– WBC (<4K or > 11K)
– Tracheal secretion (absent, thin, purulent)
– Oxygen (P/F <240 with normal PAWP: 2 points)
– CXR: presence of infiltrate (diffuse = 1; localized = 2)
– CXR: progression of infiltrate (yes = 2)
– Pathogenic bacteria on C/S (moderate = 1, same as
gram stain = 2)
Singh, Am J Respir Crit Care Med 162:505 (2000)
Revising Antibiotics Regimen
CPIS
Singh, Am J Respir Crit Care Med 162:505 (2000)
Last Week, in my ICU
• By 72 hours:
– temperature, BP, WBC normalized,
oxygenation improved but not to
baseline LLL infiltrated persisted.
baseline,
persisted
• MiniMini-BAL disclosed > 104 E. coli sensitive
to pip/tazo
pip/tazo;; blood cultures remained
negative
• Vanc
Vanc,, tigecycline and colistin discontinued.
16
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Y
Choose Drug
Safe/
N
Effective
N
Non-Non
N On the
Formulary
Formulary
Req.
N
Y
Y
Y
Restricted?
Justified?
N
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Y
N
Best
Practice?
Y
Right
N
Frequency/
Dose
Prescribe Drug
Is the Patient Cured?
VAP
•
•
•
•
Chastre et al. JAMA 290:2588 (2003)
Prospective randomized, double blind until day 8
401 VAP patients, 51 French ICUs
Exclusion Criteria:
– Immunocompromised
– Inappropriate initial empiricism
• 8 days non
non--inferior to 15 days
• Caution in P. aeruginosa (increased relapse rate)
Is the Patient Cured?
Sepsis Marker: procalcitonin
Bouadma, Lancet, 375, 463 (2010)
17
DeLisle – Local antibiotic susceptibility data - USA
Last Thursday, in my ICU
• Plan is for pip/tazo
pip/tazo to be continued for a
total of 8 days
Decisional Role of the Antibiogram
• The antibiogram has broad role to play in
decisions about antibiotics…
N
Drug
Needed?
Stop
Y
Y
Choose Drug
N
Safe/
Effective
N
NonNonN On the
Formulary
Formulary
Req.
N
Justified?
Y
Y
Y
Restricted?
N
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Y
N
Best
Practice?
Y
Right
N
Frequency/
Dose
Prescribe Drug
18
DeLisle – Local antibiotic susceptibility data - USA
Current Role of Aggregated
Antibiogram
• Development of local formularies
• Development of local guidelines for initial
p , screening
g and isolation p
procedures
therapies,
• Phased information to revise therapies
• Support recommendations from antibiotics
management team
I have seen the future…
19
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Y
Safe/
N
Effective
Choose Drug
N
Non-Non
N On the
Formulary
Formulary
Req.
N
Y
Evaluate Drug
Y
Y
Ri
Right
ht
Patient/
Time
Y
Restricted?
Justified?
Deliver Drug
N
Y
N
Y
N
Best
Practice?
Right
N
Frequency/
Dose
Y
Prescribe Drug
N
Drug
Needed?
Stop
Y
Choose Drug
N
Best
Practice
Rx?
Evaluate Drug
Y
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Antibiotics Computer Assistant
• Antibiotic use in a trauma ICU compared with and
without computer support (7/92(7/92-6/94 vs. 7/94
7/94--6/95)
Variable
Preintervention
Duration of
214
therapy (hrs)
Antibiotic
23.6
doses
Days of excess 5.4
antibiotics
Intervention
Followed
Overridden
103
330
11.4
27.6
1.4
3.6
Adjusted outcomes; differences are significant
Evans RS et al. N Engl J Med. 1998;338:232, slide courtesy of S Cosgrove
20
DeLisle – Local antibiotic susceptibility data - USA
Pharyngitis
Bronchitis
1199
1233
431
266
537
233
95
145
123
Pneumonia
Sinusitis
URI
Prescribed When Indicated for ARI
Control
Site
Prescribed When Ind
dicated
Intervention
Site
Targeted
ABX
Other
ABX
Targeted
ABX
Other
ABX
Prescribed When Not Indicated for ARI
Control
Site
Prescribed When Not In
ndicated
Intervention
Site
Targeted
ABX
Other
ABX
Targeted
ABX
Other
ABX
21
DeLisle – Local antibiotic susceptibility data - USA
N
Drug
Needed?
Stop
Y
Choose Drug
N
Best
Practice
Rx?
Y
Evaluate Drug
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
N
Drug
Needed?
Stop
Y
Choose Drug
Antibiograms
Best
Practice
Rx?
Evaluate Drug
N
Y
Y
Ri
Right
ht
Patient/
Time
N
Deliver Drug
Future Role of Aggregated
Antibiograms
• Evaluate external data
• Allow aggregation across different data
dimensions (environment, patient, site, bug)
gg g
according
g to statistical
• Guide this aggregation
principles
• Allow development and testing of algorithms that
continuously inform
– individualized sensitivity testing
– ongoing antibiotherapies
22
DeLisle – Local antibiotic susceptibility data - USA
Barriers to Implementation
• Lack of impetus and resources to do the work
• Underdeveloped health data security and
privacy
p
y statutes and exchanges
g agreements
g
• Lack of standardization for the inclusion,
conduct, thresholding,
thresholding, and reporting of
sensitivity testing
• Lack of computability of antibiogram data
elements
Summary
• Aggregated antibiograms contribute to the
evidence--based development of local
evidence
formularies, therapeutic and infection control
guidelines, and individual antibiotherapies
• Antibiogram information should accrue to
relational databases and be shared across
referral networks to support increasingly
complex decisions regarding individualized
sensitivity testing and drug therapies
3
23
Lopez-Lozano - ViResiST
www.viresist.org
Surveillance of Resistance at
Local Level by Means of Time
Series Analysis
GENERAL SCHEMA
Empirical guidelines
•
•
Most probable
microorganisms
Microbiology
Data
(Antibiograms)
Abucasis
Hospital (Intranet / Orion Clinic / Alta Hospitalaria)
Predicted
Resistance
Internet
Viresist.org
Antibiotic
Use
(DDD)
Researching
•
Antibiotic use – Resistance
Relationship
•
Ab use studies
Microbiology Data Entry
(1 of 2)
Online Culture Form
Orion Clinic
(Electronic Patient History)
Intranet Hospital /
Direct
WHONET
Microbiology
Data
(Antibiograms)
Primaria
Laboratory
Abucasis (Community)
(Electronic Patient History)
24
Lopez-Lozano - ViResiST
Micro Data
(2 of 2)
• Patient ID (NHC, SIP)
• Setting
• Antibiótic
Microbiology
Data
(Antibiograms)
Intermediary
Data
CMBD
Hospital
Demographic Data
• Crossing Data
• We get Information about::
• Sample – diagnostico – microorganism
• Antibiogram results
• Nosocomial or communitary
• Hospital Stay
Phamacy Data
CMBD
Hospital
Demographic Data
Farmasyst
(Hospital
Phamacy)
DDD/1000
pacient/days
GAIA
Stay
Antibiotic
Use
Primaria
(Community
Phamacy)
DDD/1000
Inhabitants/days
Forecasting Resistance
Microbiology
Data
(Antibiograms)
Historical database
of monthly % of resistance
(microorganism_antibiotic)
Automatic process
(adding new data)
ARIMA Models
SCA
General Data Base
Predicted
Resistance
25
Lopez-Lozano - ViResiST
Empiric Therapy Guidelines
•
•
•
More likely microorganisms infecting the
patient, ordered by decreasing expected
frequency
Expected
p
resistance of these microorganisms
g
to the available antibiotics
Recommendation of specific antibiotic
prescription considering certain characteristics
of the patient: age, sex, pregnancy or not,
renal function, weight, etc. …
26
Lopez-Lozano - ViResiST
Bacteria
name
Nosocomial
or community
origin
Strains
number
Cumulated%
Expected %
Placing the mouse over the microorganism name
Placing (without
the mouse
over the
name
clicking),
wemicroorganism
get expected resistance
…..
Escherichia
coli
Enterococcus
faecalis
27
Lopez-Lozano - ViResiST
Klebsiella
pneumoniae
28
Lopez-Lozano - ViResiST
29
Lopez-Lozano - ViResiST
Selecting
antibiotic use
We get the conjoint resistance
and antibiotic use evolution
In this case, monthly % of ESBL
producers Ecoli and Levofloxacin Use
30
Lopez-Lozano - ViResiST
Brute data
Smoothed data
31
Lopez-Lozano - ViResiST
Smoothed data
Relative over
the total ab use
32
Know your own local antibiotic susceptibility data: Asia
Po-Ren Hsueh
Divisions of Clinical Microbiology and Infectious Diseases, Departments of Laboratory
Medicine and Internal Medicine, National Taiwan University Hospital, National Taiwan
University College of Medicine, Taipei, Taiwan
Antimicrobial resistance has become a major health problem worldwide, but
marked variations in resistance profiles of bacterial pathogens are found among
countries and in different patient settings, especially in Asian countries. The emerging
problem of methicillin-resistant Staphylococcus aureus (MRSA), extended-spectrum
beta-lactamase (ESBL)-producing, quinolone-resistant, and New Delhi
Metallo-β-lactamase-1 (NDM-1)-producing Enterobacteriaceae is substantial.
Extensively drug-resistant (XDR, resistant to all antibiotics available but one or two)
Gram-negative bacilli of clinical importance include K. pneumoniae, Pseudomonas
aeruginosa, Acinetobacter baumannii, and Stenotrophomonas maltophilia. Currently,
NDM-1-producing A. baumannii also emerged in China. Current data from the Study
for Monitoring Antimicrobial Resistance Trends (SMART) demonstrated the high
resistance burden of isolates associated with intra-abdominal infections (IAIs) and
urinary tract infections (UTIs) in Asia-Pacific region. For IAI isolates, rates of
extended-spectrum β-lactamase (ESBL)-producing E. coli of >20% was found in
South Korea (25.9%), Vietnam (41.1%), and Thailand (47.9%). The rates of
susceptibility to levofloxacin of <50% was found in South Korea (48.2%) and
Thailand (45.2%). As for UTI E. coli isolates, high rates (>20%) of ESBL phenotype
was found in India (60.4%), Vietnam (60.2%), Thailand (36.8%), China (34.8%),
South Korea (25.9%), and the AP region (23.9%). Rate of ESBL-producing UTI K.
pneumoniae isolates was 50% in India and Vietnam, 46.7% in Thailand, 20% in South
Korea, and 29.5% in AP region. These up-to-date epidemiology and antimicrobial
resistance surveillance data are crucial to select appropriate treatment of IAIs and
UTIs. Appropriate use of currently available antibiotics and strict adherence of
adequate infection control policy are crucial.
33