Relationship Between Antibiotic Consumption and Resistance ... · Relationship Between Antibiotic...

STATENS SERUM INSTITUT

Relationship BetweenAntibiotic Consumption

and Resistancein European Hospitals

Dominique L. MonnetNational Center for Antimicrobials and Infection Control,

Statens Serum Institut, Copenhague, Danemark


The World (of Antimicrobial Resistance) According to… Human Bacterial Pathogens and Their Habitat

H

R Strep. pneumoniae

R Haem. influenzae

R Pseudomonas aeruginosaR Acinetobacter baumannii

Foo

dA

nim

als

Hum

ans

Hum

ans

R Salm.

R Camp.

R E. coli

R E. coli

MRSA

RS. aureus

RS. aureus

le MONDe de la Résistance Intrinsèque et Acquise aux ANtibiotiques ;-).

STATENS SERUM INSTITUTAntimicrobial Consumption and Resistance:

Examples from ARPAC European Hospitals, 2001

n=112R2 =0.07

n=110R2 =0.09

Source: ARPAC, 2004 (http://www.abdn.ac.uk/arpac/)

0 1 2 3 4 5 6 7 80

10

20

30

40

50

Carbapenem cosumption(J01DH, DDD/100 bed-days)

% Im

ipen

em-r

esis

tan

tP

seu

do

mo

nas

aer

ug

ino

sa0 5 10 15 20

0

10

20

30

40

50

Fluoroquinolone consumption(J01MA, DDD/100 bed-days)

% F

luo

roq

uin

olo

ne-

resi

stan

tE

sch

eric

hia

co

li(%

)


Antimicrobial Use (DDD/1,000 pt-days)

% R

esis

tant

bact

eria

+

++

+ +++

++

++

++

+

+

++

+

++ + +

+

Usefulness of Antimicrobial Resistance and Antimicrobial Use Data Comparison

Low level of resistanceLow antimicrobial use

High level of resistanceLow antimicrobial use

High level of resistanceHigh antimicrobial use

Low level of resistanceRelatively high antimicrobial use

Area of improvement: antimicrobial use

Possible area of improvement: detection of resistance in the laboratoryPossible explanation: resistant bacteria has not been introduced in setting

Possible areas of improvement: . infection control. identif. of colonized patientsupon admission

. appropriate dosage (low dose)

. use of other antimicrobials

Source : Int J Antimicrob Agents 2000;15:91-101 (adapted from CDC/NNIS/ICARE Phase 1).

median

median+


Source: Goossens H, et al. Lancet 1986;2:804.

R = 0.90p < 0.005

Gentamicin useprevious year (g/year)

Gentamicin usesame year (g/year)

% G

enta

mic

in-r

esis

tant

gram

-neg

ativ

e ba

cilli

Gentamicin Use and %Gentamicin-ResistantGram-Neg. Bacilli Isolates, Brussels, 1979-1986

02468

10

0 500 1000 1500 2000

02468

10

0 500 1000 1500 2000

STATENS SERUM INSTITUTPercent Ceftazidime-Resistant/Intermediate

Gram-Negative Bacilli and Hospital Ceftazidime Use, Hospital Vega Baja, Spain, 1991-1998

ViResiSTSource: Monnet DL, et al. Clin Microbiol Infect 2001; 7(Suppl 5):29-36.

Yearly data Monthly data(5-month moving average)

0

2

4

6

8

10

12

Jan-

91

Jan-

92

Jan-

93

Jan-

94

Jan-

95

Jan-

96

Jan-

97

Jan-

98

0

2

4

6

8

10

12

1991 1992 1993 1994 1995 1996 1997 1998

Ceftazidime use (DDD/1,000 pt-days) Ceftazidime-resistant GNB (%)


Source: FT Investor Financial Times, 7/29/2002.

Examples of Time SeriesDow Jones Industrial Average

Source: Aiello AE & Larson EL. Lancet Infect Dis 2002;2:103-10.

Crude Death Rates for InfectiousDiseases, USA, 1900-1996


ErrUDA with ErrRes

Nº de retardos

76543210-1-2-3-4-5-6-7

CC

F

1.0

.5

0.0

-.5

-1.0

Límites confianza

Coeficiente

CCF

n To assess relationships between a target (output) series and one orseveral explanatory (input) seriesn Various types of models: transfer

function (TF), polynomial distributedlag (PDL), etc.n TF models: cross-correlation function

(CCF) to identify time lags betweenseries

0

2

4

6

8

10

12

Jan-91

Jul-91

Jan-9

2Jul

-92Jan

-93Jul

-93Jan

-94Jul

-94Ja

n-95

Jul-95

Jan-96

Jul-96

Jan-97

Jul-97

Jan-9

8Jul

-98

0

2

4

6

8

10

12

Multivariate Time Series Analysis

Sources: Helfenstein U. Stat Meth Med Res 1996;5:3-22.Haugh LD. J Am Stat Assoc 1976;71:378-385.Pankratz A. Forecasting with dynamic regression models. New York, NY: Wiley, 1991.


Ceftazidime Use1 month before

ViResiST

R2=0.44

Average delay = 1 month+1 DDD/1,000 patient-days = 6.5 days of treatment à +0.42 %R

e.g. from R = 5% à R = 5.42 %

Source : López-Lozano JM, et al. Int J Antimicrob Agents 2000;14:21-30.

Transfer Function Model for Percent Ceftazidime-Resistant/Intermediate Gram-Negative Bacilli Series

(taking into account hospital ceftazidime use)

Term Parameter (SE) T-ratio P-value

Constant 1.354 (0.760) 1.78 0.078

AR3 0.352 (0.096) 3.68 < 0.001

AR5 0.265 (0.098) 2.72 < 0.01

ULAG1 0.420 (0.096) 4.34 < 0.0001


Source : Monnet DL, et al. Clin Microbiol Infect 2001; 7(Suppl 5):29-36. ViResiSTMonths

5-Month Moving Average Percent Amikacin-Resistant/Intermediate P. aeruginosa and Hospital

Antimicrobial Use, Hospital Vega Baja, Spain, 1991-1999

0

10

20

30

40

50

60

70

Amikacin use(DDD/1,000 pt-dys)

Amikacin-R P. aeruginosa (%)


0

10

20

30

40

50

60

70Gentamicin use(DDD/1,000 pt-dys)

Tobramycin use(DDD/1,000 pt-days)

Cefotaxime use(DDD/1,000 pt-dys)

Ceftazidime use(DDD/1,000 pt-dys)

Ceftriaxone(DDD/1,000 pt-dys)

Amikacin use(DDD/1,000 pt-dys)

Amikacin-R P. aeruginosa (%)Months

Source : Monnet DL, et al. Clin Microbiol Infect 2001; 7(Suppl 5):29-36. ViResiST

5-Month Moving Average Percent Amikacin-Resistant/Intermediate P. aeruginosa and Hospital

Antimicrobial Use, Hospital Vega Baja, Spain, 1991-1999


Source : Monnet DL, et al. Clin Microbiol Infect 2001; 7(Suppl 5):29-36. ViResiST

Transfer Function Model for PercentAmikacin-Resistant Pseudomonas aeruginosa Series

(taking into account aminoglycoside and 3rd-generation cephalosporin use)

Term Order Parameter (SE) T-ratio P-value

Constant 0 -20.741 (4.516) -4.59 < 0.001

Amikacin 7 0.973 (0.391) 2.49 < 0.02

Gentamicin 7 0.420 (0.153) 2.75 < 0.01

Cefotaxime 3 0.297 (0.112) 2.66 < 0.01

Cefotaxime 6 0.437 (0.110) 3.98 < 0.001

AR 2 0.295 (0.091) 3.24 < 0.01


* only 55.3% of isolates were tested for susceptibility to ceftriaxone

ViResiSTSource: Monnet DL, et al. Clin Microbiol Infect 2001; 7(Suppl 5):29-36.

Co-Resistances in Amikacin-Resistant/Intermediateand Susceptible Pseudomonas aeruginosa Isolates,

Hospital Vega Baja, Spain, 1991-1999

Cross-resistance Amikacin-R/I Amikacin-S RR P-valueno. (%) no. (%)

Gentamicin-R/I 78 (97.5) 177 (17.5) 128.0 <0.0000001 Cefotaxime-R/I 73 (91.3) 840 (83.0) - NS Ceftriaxone-R/I* 40 (81.6) 361 (74.7) - NS Tobramycin-R/I 34 (42.5) 18 (1.8) 14.8 <0.0000001 Ceftazidime-R/I 15 (18.8) 37 (3.7) 4.6 <0.0000001

Co-resistance


ViResiST

R2=0.902

Source: Monnet DL, et al. Emerg Infect Dis 2004;10:1432-1441.

%MRSA and Monthly Use of Macrolides, Third-Generation Cephalosporins and Fluoroquinolones,

Aberdeen Royal Infirmary, 01/1996-12/2001

Explaining variable

for monthly %MRSA

Lag (months) Estimated

coefficient

%MRSA 1 0.420

Macrolide use 1,2,3 0.165

Third-generation

cephalosporin use4,5,6,7 0.290

Fluoroquinolone use 4,5 0.255

Constant - - 36.7


ViResiSTUpdated from: López-Lozano JM, et al. Int J Antimicrob Agents 2000;14:21-30.

% Im

ipen

em-r

esis

tant

/inte

rmed

iate

Pse

udom

onas

aeru

gino

sa

Ho

spit

al im

ipen

emus

e(D

DD

/1,0

00 p

atie

nt-

day

s)

Average delay = 1 month

1 DDD/1,000 pat-days à +0.40 %R

5-Month Moving Average Percent Imipenem-Resistant/Intermediate P. aeruginosa and Hospital

Imipenem Use, Hospital Vega Baja, Spain, 1991-1999

0

5

10

15

20

25

Jan -91

Jul -9

1

Jan -92

Jul -9

2

Jan -93

Jul -9

3

Jan -94

Jul -9

4

Jan -95

Jul -9

5

Jan -96

Jul -9

6

Jan -97

Jul -9

7

Jan -98

Jul -9

80

5

10

15

20

25

30


Univ. Hospital, Ulm (D)Lepper et al. AAC 2002;46:2920-5.

Univ. Hospital, Antwerp (B)Goossens H, et al. Unpublished data.

Univ. Hospital, Utah (USA)Samore MH, et al. Unpublished data.

Centre Hosp. Mulhouse (F)Aujoulat O, Delarbre JM. ViResiST.

Averagedelay

= 0-1 month

Averagedelay

= 0-2 months

Averagedelay

= 0-1 month

Averagedelay

= n.a.

%Carbapenem-Resistant Pseudomonas aeruginosaand Carbapenem Use in 4 Hospitals, 1996-2003

ViResiST0

10

20

30

40

Jan -97

Jan -98

Jan -99

Jan -00

Jan -01

Jan -02

Car

bap

enem

use

(D

DD

/100

pt-

days

)

0

1

2

3

4

5

Car

bap

enem

-res

ista

nt

P.a

eru

gin

osa

(%)

0

10

20

30

40

Jan -96

Jan -97

Jan -98

Jan -99

Jan -00

Jan -01

Car

bap

enem

use

(D

DD

/100

pt-

days

)

0

0.5

1

1.5

2

Car

bap

enem

-res

ista

nt

P.a

eru

gin

osa

(%)

0

10

20

30

40

Jan -99

Jan -00

Jan -01

Jan -02

Jan -03

Jan -04

Car

bap

enem

use

(D

DD

/100

pt-

days

)

0

0.1

0.2

0.3

0.4

0.5

Imip

enem

-res

ista

nt

P.a

eru

gin

osa

(%)


ACR Chart

Source: Muller A, et al. (available free-of-charge, September 2005)

STATENS SERUM INSTITUTEffects of reduction of quinolone use

on antibiotic susceptibility in P. aeruginosa, Pittsburgh (PA), 2001-2004

Source: Paterson DL, et al. 44th ICAAC, Washington (DC), 30-10/2-11-2004, abstr. K-347.


Effect of Restricting Fluoroquinolones, ICU, Saint-Etienne (F), 2000-2002

Source: Aubert G, et al. J Hosp Infect 2005;59:83-89.

0

100

200

300

400

500

Jan-Dec2000

Jan-Jun2001

Jul 2001-Jun 2002

Flu

oroq

uino

lone

con

sum

ptio

n(D

DD

/1,0

00 p

atie

nt-d

ays)

0

20

40

60

80

Res

ista

nce

(%) Fluoroquinolone

consumptionCiprofloxacin-RP. aeruginosaOfloxacin-R S. aureusMRSA

STATENS SERUM INSTITUTAntibiotic Rotation and Development of Gram-Negative Antibiotic Resistance, Surgical ICU, Utrecht (NL), 2001-2002

Levofloxacin Cefpirome Levofloxacin Pip/Tazo

Source: van Loon HJ, et al. AJRCCM, in press (published online, October 29, 2004).

Proportion of patients treated (%)

Levofloxacin 40 0 52 5 Cefpirome 0 44 0 0Pip/Tazo 6 1 1 55


Cycling Mixing

180016001400120010008006004002000

0.7

0.6

0.5

0.4

0.3

0.2

0.1

Fra

ctio

n R

esis

tant

180016001400120010008006004002000

0.7

0.6

0.5

0.4

0.3

0.2

0.1180016001400120010008006004002000

0.7

0.64

0.58

0.52

0.46

0.4

0.34

0.28

0.22

0.16

0.1

Time (days)

10 day cycles 360 day cycles90 day cycles

Source: Bergstrom CT, et al. Proc Natl Acad Sci USA 2004;101:13285-90.

Effect of Cycle Length


Areas for Future Research

nAdequation between studies at patient level and time series analyses?nAre these relationships found in every hospital?nMore on the effect of interventions aiming at

rationalizing antimicrobial prescriptionsnShort cycling vs. optimal mixing of prescriptionsnMRSA vs. antimicrobial consumptionnOutbreaks vs. endemic situationsn Interaction between infection control and

antimicrobial consumption


Colistin

Pan-resistantGram-neg. bact.

Pan-Resistant Gram-Negative Bacilli

• ICU, Henry Dunant Hosp., Athens, Greece, 2001-2004Falagas ME, et al. BMC Infect Dis 2005;5:24.

0

1

2

Jan.

2001

Jan.

2002

Jan.

2003

Jan.

2004

No.

cas

esPrior colisitin use (days) 0 23 11 33,31 6

† †

• Hosp. Clinico San Carlos, Madrid, 08/2003-08/2004: >20 pts with carbapenem-R, colistin-R P. aeruginosaSánchez A, et al. Rev Esp Quimioterap 2004;17:336-40.

Carbapenem-R, colistin-S only

Gram-neg. bact.

3rd-gen. cephs-RGram-neg. bact.

Carbapenems


Illustration: Prittie EJ. Philadeplphia, PA: JC Winston, 1930.

It’s a numbers

game!

Relationship Between Antibiotic Consumption and Resistance ... · Relationship Between Antibiotic...

Documents

Transcript of Relationship Between Antibiotic Consumption and Resistance ... · Relationship Between Antibiotic...