2019 Dược lý lâm sàng kháng sinh đh dược Hà Nội

Another way to express the pharmacodynamic properties is to plot the rate of kill derived from the kill curve experiments just described as a function of concentration Mouton and Vinks

DƯỢC LÝ LÂM SÀNG TRONG

SỬ DỤNG KHÁNG SINH

Bộ môn Dược lực Trường Đại học Dược Hà Nội

Mục tiêu học tập

1 Giải thích được các bước tiếp cận hệ thống trong

lựa chọn kháng sinh

2 Thiết kế được chế độ liều trong sử dụng các kháng

sinh nhóm betalactam, aminoglycosid và

fluorquinolon dựa trên các dữ liệu dược động học

và dược lực học

3 Phân tích được các giải pháp để hạn chế đề kháng

kháng sinh

Nguyên lý chung trong điều trị nhiễm khuẩn

Chẩn đoán nhiễm khuẩn

– Sốt > 37oC

– Dấu hiệu và triệu chứng…

• WBC: tăng, hiếm khi quá 30.000-40.000 tb/mm3

(bình thường 4000-10000 tb/mm3)

• Dấu hiệu tại vị trí nhiễm khuẩn

Test đánh giá nhạy cảm

Bán định lượng

Test đánh giá nhạy cảm

Đĩa khuếch tán

Test đánh giá nhạy cảm

Xác định MIC trên đĩa 96 giếng

Test đánh giá nhạy cảm

Xác định MIC trên đĩa 96 giếng

Test đánh giá nhạy cảm

Epsilometer test (Etest)

Nguyên lý chung trong điều trị nhiễm khuẩn

"HIT HARD & HIT FAST“: nguyên tắc 4D

4D = chọn đúng kháng sinh theo phổ tác dụng, vị trí nhiễm khuẩn và nguy cơ nhiễm VK kháng thuốc, phối hợp kháng sinh hợp lý, liều dùng/chế độ liều phù hợp (PK/PD), xuống thang đúng cách

Denny KJ et al Expert Opin Drug Saf 2016; 15: 667-678.

Lựa chọn kháng sinh hợp lý

Vi khuẩn Kháng sinh

Người bệnh

Applied Pharmacokinetics and Pharmacodynamics, 4 th edition 2006.

Lựa chọn kháng sinh hợp lý

• Khả năng xâm nhập vào mô nhiễm khuẩn

• Khả năng xâm nhập vào tổ chức khác – PK/PD: AUC/MIC, C peak /MIC và T>MIC

– Độc tính của kháng sinh

Lựa chọn kháng sinh theo vi sinh

ð Điều trị kinh nghiệm

Một số b-lactam

Glycopeptid

Fluoroquinolon Tetracyclin

Sulfonamid Một số b-lactam

Nguy cơ chọn lọc

đề kháng

! Macrolid

Aminoglycosid

Lựa chọn kháng sinh theo vi sinh

Kìm hãm sự phát triển vi khuẩn

Tiêu diệt vi khuẩn

Telithromycin vs S aureus Moxifloxacin vs S aureus

MIC

MIC Nồng độ

đỉnh

Nồng độ đỉnh

Seral et al, AAC (2003) 47:228 3-2292

Bệnh nhân suy giảm miễn dịch

!

Macrolid Tetracyclin

Fluoroquinolones Aminoglycosides b-lactams

Lựa chọn kháng sinh theo vi sinh

Lựa chọn kháng sinh dựa trên đặc điểm vi sinh

thuộc vi khuẩn nghi ngờ gây bệnh)

thấp nhất trên đa số vi khuẩn

Dược lực học: ảnh hưởng của thời gian Tất cả các kháng sinh đều phụ thuộc thời gian

killing

Nhưng một số kháng sinh có tác dụng diệt khuẩn quá nhanh làm

cho thời gian không còn quan trọng

(tobramycin), hoặc

quinolon

(ciprofloxacin) tại nồng độ 4 X MIC, khả năng làm giảm

4 log số lượng vi khuẩn có thể đạt

sau 4-6h

killing

Dược lực học: ảnh hưởng của thời gian

Dược lực học: ảnh hưởng của thời gian

Dược lực học: tích hợp nồng độ và thời gian

liều- đáp ứng của thời gian lâm sàng

• Nồng độ cao không quan trọng

rộng hạn chế

• Nồng độ đóng vai trò quyết định

• Thời gian không

là yếu tố ảnh hưởng

Lựa chọn kháng sinh theo PK-PD

Liều

dùng

Hiệu quả

Độc tính

Dược lực học (Pharmacodynamics)

Liều dùng Nồng độ KS trong máu biến thiên theo

thời gian

Nồng độ KS tại vị trí nhiễm khuẩn

Nồng độ KS tại các cơ quan khác

Hiệu quả điều trị

Tác dụng phụ/độc tính

Dược động học

Tối ưu hóa theo PK/PD

Kháng sinh phụ thuộc thời gian,

không có PAE

Kháng sinh phụ thuộc nồng độ,

PAE kéo dài

Nguồn: Rybak MJ Am J Med, 2006; 119 (6A): S37-44

Phân loại kháng sinh theo PK/PD

Ứng dụng của PK/PD

• Phát triển kháng sinh mới và dạng bào chế mới

• Hướng dẫn điều trị theo kinh nghiệm

• Xác định điểm gãy nhạy cảm

Fundamentals of Antimicrobial Pharmacokinetics and Pharmacodynamics

ISBN 978-0-387-75612-7 ISBN 978-0-387-75613-4 (eBook) DOI 10.1007/978-0-387-75613-4

Springer New York Heidelberg Dordrecht London Library of Congress Control Number: 2013953328 © Springer Science+Business Media New York 2014 This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifi cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfi lms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifi cally for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer

Permissions for use may be obtained through RightsLink at the Copyright Clearance Center Violations are liable to prosecution under the respective Copyright Law

The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specifi c statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use

While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made The publisher makes no warranty, express or implied, with respect to the material contained herein

Printed on acid-free paper Springer is part of Springer Science+Business Media ( www.springer.com )

Editors

Alexander A Vinks Division of Clinical Pharmacology Cincinnati Children’s Hospital Medical Center and Department of Pediatrics

University of Cincinnati College of Medicine Cincinnati , OH , USA

Johan W Mouton Department of Medical Microbiology Radboudumc, Radboud University Nijmegen Nijmegen, The Netherlands

Hartmut Derendorf Department of Pharmaceutics University of Florida

Gainesville College of Pharmacy Gainesville , FL , USA

58

the agent will be most used For example, patients in intensive care units (ICU) generally have different pharmacokinetics with a higher volume of distribution and lower clearance than most other patients The use of different pharmacokinetic parameters in the simulations will obviously result in different conclusions with respect to the breakpoints, as was shown in case studies for ceftazidime (Mouton

et al 2005 ) and for other agents (Roberts et al 2009 ) MCS was performed using pharmacokinetic parameters from three different populations, human volunteers, patients with cystic fi brosis, and patients from the ICU In each population the derived breakpoints would have been different On the other hand, Muller et al recently showed that the results of MCS based on volunteer data obtained from phase 1 studies matched actual target attainments in phase 3 studies (Muller et al

2013 ) for ceftobiprole

The entire process as described in this chapter can be summarized in a fl ow gram as depicted in Fig 3.4 The diagram represents the different elements as recently described by the EUCAST and includes both the steps as required for new agents as well as those for established drugs (Mouton et al 2012 ) It should be borne

dia-in mdia-ind that breakpodia-ints are not set dia-in stone and that they are dependent on multiple factors Should one of these factors change, then the breakpoint should be reconsid- ered and possibly be changed if necessary The iterative process of optimizing dos- ing regimens and setting breakpoints continues after the breakpoint has been established.

Correlation Exposure -Effect

Preclinical PK/PD studies Clinical PK/PD studies

Correlation Exposure -Effect

PD target

Qualitative relationship (pk/pd index)

Quantitative relationship (value pk/pd index)

PD target Clinical Dosing Regimen

Monte Carlo Simulations Initial PK/PD breakpoint PK/PD breakpoint

MIC distributions

MCS robustness Target population Dose adjustments

Fig 3.4 Summary of the process of setting PK/PD breakpoints by EUCAST (Mouton et al 2012 )

J.W Mouton

58

the agent will be most used For example, patients in intensive care units (ICU) generally have different pharmacokinetics with a higher volume of distribution and lower clearance than most other patients The use of different pharmacokinetic parameters in the simulations will obviously result in different conclusions with respect to the breakpoints, as was shown in case studies for ceftazidime (Mouton

et al 2005 ) and for other agents (Roberts et al 2009 ) MCS was performed using pharmacokinetic parameters from three different populations, human volunteers, patients with cystic fi brosis, and patients from the ICU In each population the derived breakpoints would have been different On the other hand, Muller et al recently showed that the results of MCS based on volunteer data obtained from phase 1 studies matched actual target attainments in phase 3 studies (Muller et al

Correlation Exposure -Effect

PD target

Qualitative relationship (pk/pd index)

Quantitative relationship (value pk/pd index)

PD target

Clinical Dosing Regimen

Monte Carlo Simulations

Initial PK/PD breakpoint

PK/PD breakpoint

MIC distributions

MCS robustness Target population Dose adjustments

Fig 3.4 Summary of the process of setting PK/PD breakpoints by EUCAST (Mouton et al 2012 )

J.W Mouton

58

the agent will be most used For example, patients in intensive care units (ICU)

generally have different pharmacokinetics with a higher volume of distribution and

lower clearance than most other patients The use of different pharmacokinetic

parameters in the simulations will obviously result in different conclusions with

respect to the breakpoints, as was shown in case studies for ceftazidime (Mouton

et al 2005 ) and for other agents (Roberts et al 2009 ) MCS was performed using

pharmacokinetic parameters from three different populations, human volunteers,

patients with cystic fi brosis, and patients from the ICU In each population the

derived breakpoints would have been different On the other hand, Muller et al

recently showed that the results of MCS based on volunteer data obtained from

phase 1 studies matched actual target attainments in phase 3 studies (Muller et al

2013 ) for ceftobiprole

The entire process as described in this chapter can be summarized in a fl ow

dia-gram as depicted in Fig 3.4 The diagram represents the different elements as

recently described by the EUCAST and includes both the steps as required for new

agents as well as those for established drugs (Mouton et al 2012 ) It should be borne

in mind that breakpoints are not set in stone and that they are dependent on multiple

factors Should one of these factors change, then the breakpoint should be

reconsid-ered and possibly be changed if necessary The iterative process of optimizing

dos-ing regimens and settdos-ing breakpoints continues after the breakpoint has been

established.

Correlation Exposure -Effect

Preclinical PK/PD studies Clinical PK/PD studies

Correlation Exposure -Effect

PD target

Qualitative relationship (pk/pd index)

Quantitative relationship (value pk/pd index)

PD target

Clinical Dosing Regimen

Monte Carlo Simulations

Initial PK/PD breakpoint

PK/PD breakpoint

MIC distributions

MCS robustness Target population

Dose adjustments

Fig 3.4 Summary of the process of setting PK/PD breakpoints by EUCAST (Mouton et al 2012 )

J.W Mouton

Tối ưu hóa chế độ liều theo PK/PD

*Phụ thuộc T>MIC *Phụ thuộc AUC/MIC *Phụ thuộc AUC/MIC và peak/MIC65

(i.e skin infection), it is primarily the unbound fraction of drugs that crosses the membrane to the infected tissues such as the subcutaneous adipose tissues, skin, or skeletal muscles An advanced methodology to overcome such problem is to utilize microdialysis as a technique to determine the free fraction of drug exposure at the site of infection An example of implementing this methodology in the clinical setting is shown in Fig 4.1 , where the profiles of unbound ceftobiprole concentra-

tions in different tissues are presented (Barbour et al 2009b ) Note that due to different unbound drug concentrations observed in plasma versus infected sites, an unoptimized dosing scheme could be proposed based on the total plasma drug profile alone, instead of the ideal scenario which is designed based on the free drug concentration.

Thirdly, the MIC-based PKPD modeling also rely on limited PD information

The single time point of MIC is empirical and assumes that it is stationary The MIC value is laboratory dependent; dilution factors, laboratory condition, and techni- cian’s interpretation of what constitutes no growth can contribute to the inter- laboratory variability The rate of bactericidal or bacteriostatic effect with changing drug concentrations is also unknown from such simplified approach Multiple killing patterns can converge to the same MIC when only one time point is measured

Table 4.1 Pattern of MIC-based PKPD index Ambrose et al., Clin Inf Dis 44:79 (2007 ) Antimicrobial agent Bactericidal pattern of in vitro activity PK–PD measure(s) Aminoglycosides Concentration dependent AUC0–24:MIC, Cmax:MIC

β-lactams

Penicillins Time dependent T > MIC Cephalosporins Time dependent T > MIC Carbapenems Time dependent T > MIC Monobactams Time dependent T > MIC

Glycopeptides/lipopeptides Daptomycin Concentration dependent AUC0–24:MIC, Cmax:MIC Oritavancin Concentration dependent T > MIC, Cmax:MIC

Macrolides and clindamycin Azithromycin Time dependent AUC 0–24 :MIC Clarithromycin Time dependent AUC0–24:MIC Teilithromycin Concentration dependent AUC0–24:MIC Metronidazole Concentration dependent AUC0–24:MIC, Cmax:MIC Oxazolidinones

Quinolones Concentration dependent AUC0–24:MIC, Cmax:MIC Tetracyclines

Doxycyeline Time dependent AUC0–24:MIC Tigecycline Time dependent AUC0–24:MIC

Note: AUC 0–24 :MIC, the ratio of the area under the concentration–time curve at 24 h to the MIC;

Cmax:MIC, the ratio of the maximal drug concentration to the MIC; T > MIC, duration of time a

drug concentration remains above the MIC

4 Principles of Applied Pharmacokinetic–Pharmacodynamic Modeling 65

(i.e skin infection), it is primarily the unbound fraction of drugs that crosses the membrane to the infected tissues such as the subcutaneous adipose tissues, skin, or skeletal muscles An advanced methodology to overcome such problem is to utilize microdialysis as a technique to determine the free fraction of drug exposure at the site of infection An example of implementing this methodology in the clinical setting is shown in Fig 4.1 , where the profiles of unbound ceftobiprole concentra-

tions in different tissues are presented (Barbour et al 2009b ) Note that due to different unbound drug concentrations observed in plasma versus infected sites, an unoptimized dosing scheme could be proposed based on the total plasma drug profile alone, instead of the ideal scenario which is designed based on the free drug concentration.

Thirdly, the MIC-based PKPD modeling also rely on limited PD information The single time point of MIC is empirical and assumes that it is stationary The MIC value is laboratory dependent; dilution factors, laboratory condition, and techni- cian’s interpretation of what constitutes no growth can contribute to the inter- laboratory variability The rate of bactericidal or bacteriostatic effect with changing drug concentrations is also unknown from such simplified approach Multiple killing patterns can converge to the same MIC when only one time point is measured

Table 4.1 Pattern of MIC-based PKPD index Ambrose et al., Clin Inf Dis 44:79 (2007)

Antimicrobial agent Bactericidal pattern of in vitro activity PK–PD measure(s)

β-lactams

Glycopeptides/lipopeptides

Macrolides and clindamycin

Oxazolidinones

Tetracyclines

Note: AUC0–24:MIC, the ratio of the area under the concentration–time curve at 24 h to the MIC;

Cmax:MIC, the ratio of the maximal drug concentration to the MIC; T > MIC, duration of time a

drug concentration remains above the MIC

4 Principles of Applied Pharmacokinetic–Pharmacodynamic Modeling

Tối ưu hóa chế độ liều theo PK/PD

hoặc ngắn

nhiễm với thuốc

et al 1996; Mouton and Vinks 2005b) Slight differences in degree and rate of

killing may exist between penicillins, cephalosporins and carbapenems, with

carbapenems being most rapidly bactericidal and penicillins least against

Gram-negative pathogens (Periti and Nicoletti 1999) In contrast to the beta-lactams,

several other antibiotics, including aminoglycosides, show a clear

concentration-dependent killing, in that killing of bacteria increases with increasing concentration

(Vogelman and Craig 1986).

Another way to express the pharmacodynamic properties is to plot the rate of kill

derived from the kill curve experiments just described as a function of concentration

(Mouton and Vinks 2005a) This is shown in Fig 10.2 (Mouton and Vinks 2005a, b)

Here from, it can be concluded that the maximum kill rate of ceftazidime and

meropenem are reached at around four times the MIC Since the maximum killing

effect of beta-lactams is reached at four times the MIC and higher concentrations

not further contributing to the increase of the antimicrobial effect, the postulate was

and is that high peak concentrations after intermittent infusion do not contribute to

efficacy, whereas prolonged concentrations below the MIC result in reduced

effi-cacy In contrast, continuous administration resulting in concentrations above the

MIC, but preferably above four times the MIC during the whole dosing interval,

should result in prolonged activity In a simulation study, we demonstrated that

efficacy is maximised when free drug concentrations are maintained at

concentra-tions that result in maximum bactericidal activity, thus four times the MIC (Mouton

and Vinks 2005b).

Fig 10.2 Relationship between concentration of ceftazidime (a) and meropenem (b) and kill rate

The relationship follows a Hill type model with a relatively steep curve; the difference between no

effect (growth, here displayed as a negative kill rate) and maximum effect is within 2–3 twofold

dilutions The maximum kill rate is attained at around 4× MIC Figure modified from Mouton and

Vinks (2005b, 2007) Reproduced from Mouton JW, Vinks AA

Pharmacokinetic/pharmacody-namic modelling of antibacterials in vitro and in vivo using bacterial growth and kill kinetics: the

minimum inhibitory concentration versus stationary concentration Clin Pharmacokinet

2005;44(2):201–10 with permission from Adis (© Springer International Publishing AG [2005]

All rights reserved

A.E Muller and J.W Mouton

Tối ưu hóa chế độ liều betalactam theo PK/PD

• Dược lực betalactam in vitro

226

et al 1996 ; Mouton and Vinks 2005b ) Slight differences in degree and rate of killing may exist between penicillins, cephalosporins and carbapenems, with carbapenems being most rapidly bactericidal and penicillins least against Gram- negative pathogens (Periti and Nicoletti 1999 ) In contrast to the beta-lactams, several other antibiotics, including aminoglycosides, show a clear concentration- dependent killing, in that killing of bacteria increases with increasing concentration (Vogelman and Craig 1986 ).

Another way to express the pharmacodynamic properties is to plot the rate of kill derived from the kill curve experiments just described as a function of concentration (Mouton and Vinks 2005a ) This is shown in Fig 10.2 (Mouton and Vinks 2005a , )

Here from, it can be concluded that the maximum kill rate of ceftazidime and meropenem are reached at around four times the MIC Since the maximum killing effect of beta-lactams is reached at four times the MIC and higher concentrations not further contributing to the increase of the antimicrobial effect, the postulate was and is that high peak concentrations after intermittent infusion do not contribute to efficacy, whereas prolonged concentrations below the MIC result in reduced effi- cacy In contrast, continuous administration resulting in concentrations above the MIC, but preferably above four times the MIC during the whole dosing interval, should result in prolonged activity In a simulation study, we demonstrated that efficacy is maximised when free drug concentrations are maintained at concentra- tions that result in maximum bactericidal activity, thus four times the MIC (Mouton and Vinks 2005b ).

Fig 10.2 Relationship between concentration of ceftazidime (a) and meropenem (b) and kill rate

The relationship follows a Hill type model with a relatively steep curve; the difference between no effect (growth, here displayed as a negative kill rate) and maximum effect is within 2–3 twofold dilutions The maximum kill rate is attained at around 4× MIC Figure modified from Mouton and Vinks ( 2005b , 2007 ) Reproduced from Mouton JW, Vinks AA Pharmacokinetic/pharmacody- namic modelling of antibacterials in vitro and in vivo using bacterial growth and kill kinetics: the minimum inhibitory concentration versus stationary concentration Clin Pharmacokinet

2005;44(2):201–10 with permission from Adis (© Springer International Publishing AG [2005]

All rights reserved

A.E Muller and J.W Mouton

226

et al 1996 ; Mouton and Vinks 2005b ) Slight differences in degree and rate of killing may exist between penicillins, cephalosporins and carbapenems, with carbapenems being most rapidly bactericidal and penicillins least against Gram- negative pathogens (Periti and Nicoletti 1999 ) In contrast to the beta-lactams, several other antibiotics, including aminoglycosides, show a clear concentration- dependent killing, in that killing of bacteria increases with increasing concentration (Vogelman and Craig 1986 ).

Another way to express the pharmacodynamic properties is to plot the rate of kill derived from the kill curve experiments just described as a function of concentration (Mouton and Vinks 2005a ) This is shown in Fig 10.2 (Mouton and Vinks 2005a , b ) Here from, it can be concluded that the maximum kill rate of ceftazidime and meropenem are reached at around four times the MIC Since the maximum killing effect of beta-lactams is reached at four times the MIC and higher concentrations not further contributing to the increase of the antimicrobial effect, the postulate was and is that high peak concentrations after intermittent infusion do not contribute to efficacy, whereas prolonged concentrations below the MIC result in reduced effi- cacy In contrast, continuous administration resulting in concentrations above the MIC, but preferably above four times the MIC during the whole dosing interval, should result in prolonged activity In a simulation study, we demonstrated that efficacy is maximised when free drug concentrations are maintained at concentra- tions that result in maximum bactericidal activity, thus four times the MIC (Mouton and Vinks 2005b ).

Fig 10.2 Relationship between concentration of ceftazidime (a) and meropenem (b) and kill rate

The relationship follows a Hill type model with a relatively steep curve; the difference between no effect (growth, here displayed as a negative kill rate) and maximum effect is within 2–3 twofold dilutions The maximum kill rate is attained at around 4× MIC Figure modified from Mouton and Vinks ( 2005b , 2007 ) Reproduced from Mouton JW, Vinks AA Pharmacokinetic/pharmacody- namic modelling of antibacterials in vitro and in vivo using bacterial growth and kill kinetics: the minimum inhibitory concentration versus stationary concentration Clin Pharmacokinet 2005;44(2):201–10 with permission from Adis (© Springer International Publishing AG [2005] All rights reserved

A.E Muller and J.W Mouton

226

et al 1996; Mouton and Vinks 2005b) Slight differences in degree and rate of

killing may exist between penicillins, cephalosporins and carbapenems, with

carbapenems being most rapidly bactericidal and penicillins least against

Gram-negative pathogens (Periti and Nicoletti 1999) In contrast to the beta-lactams,

several other antibiotics, including aminoglycosides, show a clear

concentration-dependent killing, in that killing of bacteria increases with increasing concentration

(Vogelman and Craig 1986)

Another way to express the pharmacodynamic properties is to plot the rate of kill derived from the kill curve experiments just described as a function of concentration

(Mouton and Vinks 2005a) This is shown in Fig 10.2 (Mouton and Vinks 2005a, b)

Here from, it can be concluded that the maximum kill rate of ceftazidime and

meropenem are reached at around four times the MIC Since the maximum killing

effect of beta-lactams is reached at four times the MIC and higher concentrations

not further contributing to the increase of the antimicrobial effect, the postulate was

and is that high peak concentrations after intermittent infusion do not contribute to

efficacy, whereas prolonged concentrations below the MIC result in reduced

effi-cacy In contrast, continuous administration resulting in concentrations above the

MIC, but preferably above four times the MIC during the whole dosing interval,

should result in prolonged activity In a simulation study, we demonstrated that

efficacy is maximised when free drug concentrations are maintained at

concentra-tions that result in maximum bactericidal activity, thus four times the MIC (Mouton

and Vinks 2005b)

Fig 10.2 Relationship between concentration of ceftazidime (a) and meropenem (b) and kill rate

The relationship follows a Hill type model with a relatively steep curve; the difference between no

effect (growth, here displayed as a negative kill rate) and maximum effect is within 2–3 twofold

dilutions The maximum kill rate is attained at around 4× MIC Figure modified from Mouton and

Vinks ( 2005b , 2007 ) Reproduced from Mouton JW, Vinks AA

Pharmacokinetic/pharmacody-namic modelling of antibacterials in vitro and in vivo using bacterial growth and kill kinetics: the

minimum inhibitory concentration versus stationary concentration Clin Pharmacokinet

2005;44(2):201–10 with permission from Adis (© Springer International Publishing AG [2005]

All rights reserved

A.E Muller and J.W Mouton

Tối ưu hóa chế độ liều betalactam theo PK/PD

• Thông số PK/PD của ceftazidim trên P.aeruginosa (mô hình gây

nhiễm khuẩn trên chuột)

Tối ưu hóa chế độ liều betalactam theo PK/PD

• Thông số PK/PD của imipenem trên P.aeruginosa (mô hình gây

nhiễm khuẩn trên chuột nhắt trắng)

6

cephalosporins, carbapenems, and monobactams With this pattern, one would

predict that the duration of time that active antibiotic concentrations exceeded the

MIC would be the important PK/PD index for effi cacy Figure 1.2 demonstrates the

relationships among the various PK/PD indices for total drug concentration of

imipenem, a carbapenem ß-lactam antibiotic with protein binding <5 % in mice,

against a standard strain of Pseudomonas aeruginosa in the thighs of neutropenic

mice The percentage of the dosing interval that concentrations exceeded the MIC

showed the best correlation with organism growth and killing, while the

relation-ships with AUC/MIC and peak/MIC looked more like scattergrams.

The third pattern of antimicrobial activity also exhibits concentration- independent

killing but these antimicrobials induce prolonged persistent effects This pattern is

observed with a large number of antimicrobials including the tetracyclines,

tigecy-cline, macrolides, azithromycin, clindamycin, linezolid and other oxazolidinones,

chloramphenicol, trimethoprim, sulfonamides, vancomycin, and dalbavancin

Because the prolonged persistent effects will protect against regrowth when active

drug concentration fall below the MIC, one would predict that the amount of drug

or the AUC/MIC would be the important PK/PD index for these drugs Figure 1.3

illustrates that relationship between the change in effi cacy from the start of therapy

and the various PK/PD indices based on total drug concentrations for vancomycin

(protein binding 13 % in mice) (Rybak 2006 ) The best correlation for effi cacy was

seen with 24-h AUC/MIC index Peak/MIC and time above MIC showed much

more variation in effi cacy at different magnitudes of the index.

Magnitude of Index Required for Effi cacy

Once the important PK/PD index driving effi cacy is identifi ed, the next piece of

information needed is what magnitude of the index is required for antimicrobial

effi cacy A large number of animal studies on the effi cacy of ß-lactams against

24 Hour AUC/MIC

Peak/MIC

1 10 100 100010000

Fig 1.2 Relationship between three PK/PD indices for total drug of imipenem and the log 10 CFU/

thigh over 24 h for Pseudomonas aeruginosa ATCC 27853 in the thighs of neutropenic mice

W.A Craig

5

amikacin from 18 to 110 min by inducing renal impairment also enhanced the AUC about sixfold, but the longer duration of sub-MIC concentrations increased the in vivo postantibiotic effect from 7.4 to 12.2 h The role of leukocytes on the in vivo PAE has also been assessed Studies with similar doses of gentamicin against the

same strain of K pneumoniae have reported in vivo PAEs of 7.8, 12.0, and 16.5 h in

neutropenic, normal, and granulocytic mice, respectively (Shimizu et al 1989 )

Patterns of Antimicrobial Activity

Three major patterns of antimicrobial activity have been observed The fi rst applies to antimicrobials with concentration-dependent killing along with prolonged persistent effects This pattern is observed with aminoglycosides, fl uoroquinolones, polymyxins, daptomycin, and some of the new glycopeptides, such as telavancin and oritavancin, which also exhibit an additional membrane effect mechanism of action One would predict that the ratio of the AUC and peak concentration to the MIC would be the pri- mary PK/PD indices correlating with antimicrobial effi cacy Done- fractionation stud- ies in animal models of infection in which fi ve or six total doses are divided into many smaller doses given at different dosing frequencies have been useful in reducing the interdependence among the PK/PD indices and confi rming which PK/PD index is most important for effi cacy The relationship of all the PK/PD indices based on total drug concentrations (protein binding in mice 15 %) to effi cacy of levofl oxacin against

Streptococcus pneumoniae in the thighs of neutropenic mice are shown in Fig 1.1 (Andes and Craig 2002 ) The 24-h AUC/MIC showed the best correlation for effi cacy followed by the peak/MIC ratio The time above MIC looked more like a scattergram The second pattern of antimicrobial activity is the exact opposite of the fi rst pat- tern with concentration-independent killing and no or very short persistent effects This pattern is characteristic of all of the ß-lactam antibiotics, such as penicillins,

Peak/MIC

1 10 100 1000

Time Above MIC

0 25 50 75 100

Fig 1.1 Relationship between three PK/PD indices for total drug of levofl oxacin and the log 10

CFU/thigh at 24 h for Streptococcus pneumoniae ATCC 10813 in the thighs of neutropenic mice

Reproduced with permission from Andes and Craig ( 2002 )

1 Introduction to Pharmacodynamics

S pneumoniae and fl uoroquinolones against Enterobacteriaceae and P aeruginosa

have evaluated different index magnitudes in various infection models using survival

as the endpoint The infections included pneumonia, peritonitis, bacteremia, and thigh-infection models Untreated or saline-treated controls had 80–100 % mortality

by the end of each study Figure 1.4 shows the relationship between various free drug time above MIC values for penicillins and cephalosporins versus survival of mice

with S pneumoniae infections (Andes and Craig 2000 ; Nicolau et al 2000 ) Ninety percent (90 %) or higher survival was observed when time above MIC was 35 % or higher Figure 1.5 illustrates the relationship between 24-h AUC/MIC values for multiple fl uoroquinolones and survival of mice, rats, and guinea pigs infected with

Enterobacteriaceae or P aeruginosa (Andes and Craig 2002 ; Craig and Dalhoff

1998 ) This time 90 % or higher survival was observed when the 24-h AUC/MIC value was 105 or higher This value is equivalent to averaging a little over four times the MIC for 24 h Survival was only 50 % when the 24-h AUC/MIC value was 41.

Peak/MIC

10 30 100 3001000

Time Above MIC

20 40 60 80 100

Fig 1.3 Relationship between three PK/PD indices for total drug of vancomycin and the change

in log 10 CFU/thigh over 24 h for Staphylococcus aureus ATCC 25923 in the thighs of neutropenic

mice Redrawn from data in Rybak ( 2006 )

Free Drug Time Above MIC (%)

0 20 40 60 80 100

Survival After 5 Days of Therapy 0

20 40 60 80

100

Fig 1.4 Relationship

between survival in neutropenic mice infected

with strains of Streptococcus

pneumoniae and time above MIC for various penicillins and cephalosporins Redrawn from data in Andes and Craig ( 2002 )

1 Introduction to Pharmacodynamics

Tối ưu hóa chế độ liều betalactam theo PK/PD

• Mối liên quan giữa tỷ lệ sống sót và T>MIC của các penicillin và

cephalosporin (MH gây nhiễm S.pneumoniae trên chuột nhắt trắng)

7

S pneumoniae and fl uoroquinolones against Enterobacteriaceae and P aeruginosa

have evaluated different index magnitudes in various infection models using survival

as the endpoint The infections included pneumonia, peritonitis, bacteremia, and

thigh-infection models Untreated or saline-treated controls had 80–100 % mortality

by the end of each study Figure 1.4 shows the relationship between various free drug

time above MIC values for penicillins and cephalosporins versus survival of mice

with S pneumoniae infections (Andes and Craig 2000 ; Nicolau et al 2000 ) Ninety

percent (90 %) or higher survival was observed when time above MIC was 35 % or

higher Figure 1.5 illustrates the relationship between 24-h AUC/MIC values for

multiple fl uoroquinolones and survival of mice, rats, and guinea pigs infected with

Enterobacteriaceae or P aeruginosa (Andes and Craig 2002 ; Craig and Dalhoff

1998 ) This time 90 % or higher survival was observed when the 24-h AUC/MIC

value was 105 or higher This value is equivalent to averaging a little over four times

the MIC for 24 h Survival was only 50 % when the 24-h AUC/MIC value was 41.

Peak/MIC

10 30 100 3001000

Time Above MIC

20 40 60 80 100

Fig 1.3 Relationship between three PK/PD indices for total drug of vancomycin and the change

in log 10 CFU/thigh over 24 h for Staphylococcus aureus ATCC 25923 in the thighs of neutropenic

mice Redrawn from data in Rybak ( 2006 )

Free Drug Time Above MIC (%)

Survival After 5 Days of Therapy 0

20406080

100

Fig 1.4 Relationship

between survival in

neutropenic mice infected

with strains of Streptococcus

pneumoniae and time above

MIC for various penicillins

and cephalosporins Redrawn

from data in Andes and Craig

( 2002 )

1 Introduction to Pharmacodynamics

Tối ưu hóa chế độ liều betalactam theo PK/PD

Nhiễm trùng nhẹ

Tối ưu hóa chế độ liều betalactam theo PK/PD

• Dữ liệu PK/PD của betalactam

Tương quan giữa T>MIC và tỷ lệ khỏi của bệnh nhi viêm tai giữa

Andes & Craig Pediatr Infect Dis J 1996