(Chest. 1999;115:19S-23S.)
© 1999
American College of Chest Physicians
Pharmacodynamic Principles of Antimicrobial Therapy in the Prevention of Resistance*
David S. Burgess, PharmD
* From the University of Texas at Austin, and the University of Texas
Health Science Center at San Antonio.
 |
Abstract
|
|---|
Pharmacodynamic properties can be used to divide antibiotics into
two major classes based on their mechanism of bactericidal action: (1)
concentration-dependent drugs, such as aminoglycosides and
fluoroquinolones, and (2) concentration-independent drugs, including
the ß-lactams. Antibiotics also differ in the postantibiotic effect
(PAE) that they exert. In general, concentration-dependent drugs have a
more prolonged PAE than concentration-independent drugs, particularly
against Gram-negative pathogens. Pharmacodynamic classifications have
important implications for the planning of drug regimens. For
concentration-dependent drugs, peak concentration to minimal inhibitory
concentration (MIC) ratios of approximately 10 are associated with
clinical success. Therefore, high drug levels should be the goal of
therapy. This is best achieved by high doses taken once daily. This
approach, however, is not feasible for the fluoroquinolones owing to
dose-limiting CNS toxicity. Concentration-independent agents are most
effective when the duration of serum concentrations is higher than the
pathogen’s MIC (time > MIC) for a significant proportion of the
dosing interval. Frequent dosing or continuous infusions can increase
the time > MIC. Concentrations of antibiotics that are sublethal can
permit the emergence of resistant pathogens. Optimization of antibiotic
regimens on the basis of pharmacodynamic principles could thus
significantly diminish the emergence of antibiotic
resistance.
Key Words: antibiotic pharmacodynamics • antibiotic pharmocokinetics • concentration-dependent drugs • concentration-independent drugs • continuous infusion • postantibiotic effect
|
Introduction
|
|---|
Many
factors can contribute to the development of bacterial resistance to
anti-infective agents, but one of the most important risk factors is
repeated exposure to suboptimal antibiotic concentrations.
Pharmacodynamic data, which describe the relationship between drug
serum concentration and the pharmacologic effects of the drug, may be
useful in designing regimens that minimize the likelihood of exposing
pathogens to sublethal drug levels. This article will review some of
the key principles of antibiotic pharmacodynamics and discuss the
implications of these data for improving antibiotic regimens and
preventing the emergence of resistant pathogens.
|
Clinical Pharmacology of Antimicrobial Therapy
|
|---|
As shown in Figure 1
,
pharmacokinetics and pharmacodynamics are interrelated. The
pharmacokinetic properties of a drug characterize the rise and fall of
drug concentrations in serum or tissue over time. Pharmacodynamic
parameters integrate the microbiological activity and pharmacokinetics
of an anti-infective drug by focusing on its biological effects, in
particular growth inhibition and killing of pathogens.
On the basis of pharmacodynamic properties, antimicrobial agents can be
divided into two major groups (Table 1
).
The first group, which includes the fluoroquinolones and the
aminoglycosides, consists of agents that exhibit
concentration-dependent killing of pathogens. For this group, the
higher the drug concentration, the faster the eradication of pathogens.
In the second group, which includes the ß-lactam antibiotics, peak
concentrations are relatively unimportant. Instead, the length of time
that concentrations are maintained above the pathogen’s minimal
inhibitory concentration (MIC) is critical to bacterial eradication.
Members of this group are referred to as concentration-independent or
time-dependent drugs.
|
Postantibiotic Effect
|
|---|
The postantibiotic effect (PAE) is the phenomenon of continued
suppression of bacterial growth after a short exposure of bacteria to
antimicrobial agents. This effect is probably the result of several
mechanisms, including nonlethal damage caused by the antibiotic and
continued persistence of the drug at the bacteria’s drug-binding site
for a time after drug is removed.
The PAE is influenced by several factors, including the
microorganism, the inoculum size, the antibiotic, the concentration of
antibiotic, and the duration of exposure. Distinctions in PAE between
Gram-positive and Gram-negative pathogens have also been observed. For
many antibiotics, including ß-lactam agents, the PAE against
Gram-positive pathogens lasts approximately 1 to 2 h.
Aminoglycosides and fluoroquinolones also demonstrate a Gram-negative
PAE of about 
2 h. However, ß-lactam drugs other than imipenem
show a negligible PAE against Gram-negative bacteria. Accordingly, once
concentrations of these drugs fall below the MIC, beneficial antibiotic
effects disappear and Gram-negative pathogens can start reproducing.
PAEs determined by in vitro methods may not always reflect
in vivo PAE.1
,2
,3
,4
,5
However, in vitro
PAEs often underestimate the duration of the PAE observed in
vivo.6
Exceptions to this, however, are PAEs for
streptococci that have been exposed to penicillins or cephalosporins.
In this system, penicillins exhibit significant PAEs against
streptococci in vitro but no PAEs in
vivo.3
,7
PAEs measured in vitro may also
incorrectly predict the effect of multiple dosing. In vitro,
the PAE of aminoglycosides is lost after multiple dosing, while
in vivo PAEs appear to continue.4
,5
|
The Correlation Between Pharmacodynamic Parameters and Clinical
Success
|
|---|
The pharmacodynamic parameters that correlate with clinical
success differ for concentration-dependent drugs vs those that are
relatively concentration independent (Table 1
). For the
concentration-dependent drugs (aminoglycosides, fluoroquinolones), the
best predictive parameter is either the ratio of peak drug
concentration to MIC or the ratio of area under the time-concentration
curve (AUC) to MIC. In a study of the clinical response to
aminoglycosides, a highly significant difference (p = 0.00001) was
found in the ratio of maximal peak drug concentrations to MIC between
patients who did not respond to therapy and those who
did.8
Further analyses indicated that as the maximal peak
to MIC ratios increased, so did the clinical response rate. An
approximately 90% response rate was achieved at ratios of 8 to 12 (Fig 2
).
8
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Relationship between the maximal peak plasma level
to MIC ratio and the rate of clinical response in 236 patients with
Gram-negative bacterial infection treated with aminoglycosides
(gentamicin, tobramycin, or amikacin). Vertical bars represent SE
values. From Moore et al,8
with permission.
|
|
These data indicate that for aminoglycosides, maximum drug
concentrations of at least 10 to 12 times the MIC are needed for the
successful treatment of pathogens. Similar results have been
demonstrated for the fluoroquinolones in a neutropenic model of
Pseudomonas sepsis.9
However, toxic side effects
associated with the fluoroquinolones prevent the attainment of high
peak concentrations in clinical settings. For this reason, many studies
have examined AUC to MIC ratios instead; this parameter is also
referred to as area under the inhibitory curve (AUIC). Increasing AUICs
have been correlated with a higher level of clinical and
microbiological cures in patients with pneumonia treated with
ciprofloxacin. The critical AUIC ratio needed for success in this case
appears to be approximately 125 times the MIC. AUIC values lower than
this are associated with microbiological cure rates of < 30%, while
AUIC values above this result in microbiological cure rates of
> 80%.10
The AUICs of a particular drug can vary
widely. For example, the MIC50 of Pseudomonas
aeruginosa to trovafloxacin is 1 µg/mL.11
Hence, the AUIC for 200 mg po qd and 300 mg IV qd would be 34 and 46,
respectively.
For concentration-independent drugs such as the ß-lactams, the key
parameter associated with clinical success is the percent of time that
drug levels at the site of infection exceed the MIC (time > MIC).
Percent of time above MIC of > 40% generally correlates with high
bacteriologic cure rates (Fig 3
).
The time above the MIC required for maximal ß-lactam activity may
differ depending on the pathogen. In animal studies, the maximal effect
of ß-lactams against Staphylococcus aureus is observed
when the time above the MIC is greater than approximately 40% of the
dosing interval. For S pneumoniae and Enterobacteriaceae,
maximal effect is seen when 60 to 70% of the dosing interval is above
the MIC.12
For example, the MIC50 of P
aeruginosa to piperacillin/tazobactam is 8.0
µg/mL.11
Hence, the time above the MIC for
piperacillin/tazobactam, 3.375 g IV q4h, would be the entire dosing
interval. Even if the MIC90, a more conservative value,
were utilized, the time above the MIC would still be 50%.
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Pharmacodynamics of ß-lactams: the relationship
between the time above MIC against S pneumoniae (open
symbols) and Haemophilus influenzae (closed symbols) and
bacteriologic cure. From Craig and Andes,18
with
permission.
|
|
Alternatively, clinical response to the ß-lactams has been shown to
correlate with AUIC; this is not surprising, as an increase in the AUIC
would also result in increased time above the MIC. Because of its
relevance to both concentration-dependent and concentration-independent
drugs, the AUIC has been proposed as a universal pharmacodynamic
parameter for assessing the clinical efficacy of all anti-infective
agents. However, ß-lactam antibacterial activity more closely
correlates with time above the MIC than with AUIC.12
|
Optimizing Antibiotic Regimens on the Basis of Pharmacodynamic
Characteristics
|
|---|
An understanding of the pharmacodynamic characteristics of an
antibiotic can provide insights into the best regimen for a given drug.
For concentration-dependent antibiotics, a high once-daily dose is the
best way to eradicate pathogens. This approach has been successful for
aminoglycosides but cannot be applied to fluoroquinolones because CNS
toxicity may result from high doses. Possible ways to optimize
fluoroquinolone regimens include the administration of larger doses
than conventionally used or the addition of a second agent to the
regimen.13
For concentration-independent drugs, the goal is to maximize the
time that drug levels at the site of infection exceed the pathogen’s
MIC. One way of achieving this is through continuous infusions, which
provides a prolonged time > MIC compared with bolus
dosing.13 The enhanced duration of the antibiotic effect
may be particularly important in immunosuppressed patients or in the
treatment of pathogens with high MICs. Continuous infusions also reduce
the amount of drug required for treatment.
Continuous infusions of ceftazidime have been studied in healthy
volunteers and in critically ill patients.14
,15
In 12
volunteers, continuous infusion regimens appeared to have advantages
over standard intermittent bolus dosing, particularly with respect to
time > MIC (Table 2
).
In addition, continuous infusion allowed a reduction in total daily
dose compared with conventional regimens.14
No adverse
effects were observed in healthy volunteers or critically ill patients
receiving continuous infusions of ceftazidime.14
,15
View this table:
[in this window]
[in a new window]
|
Table 2. Effect of Dosing Regimen on the Percentage of Dosing
Interval in Which Ceftazidime Serum Concentrations Exceed the MIC in 12
Healthy Volunteers*
|
|
Although initial results with continuous infusion regimens are
promising, several issues need to be addressed, including the tissue
concentrations achieved, adverse reactions, drug stability and
compatibility, and ideal concentration/MIC ratio. Perhaps the most
important issue, however, is efficacy. Although pharmacokinetic data
indicate that continuous infusion of ß-lactam agents should be
clinically effective, to my knowledge, efficacy has not been addressed
by the studies conducted to date.
|
Implications for Drug Resistance
|
|---|
The presence of sublethal concentrations of a drug exerts
selective pressure on a population of pathogens without eradicating it.
Under these circumstances, mutant strains that possess a degree of drug
resistance are favored and tend to dominate the population. From such
populations with low-level resistance, more highly resistant organisms
are more readily selected. It thus follows that one tactic to prevent
the emergence of resistance is to minimize the time that suboptimal
drug levels are present by thoughtful attention to
dosing.16
For fluoroquinolones and aminoglycosides, optimal dosing means
maintaining high maximum peak concentration to MIC ratios.
Pharmacodynamic data thus predict that problems with resistance may
arise when fluoroquinolones are used to treat pathogens with high
MIC values, such as Pseudomonas, because the highest drug levels
attained are only approximately four to five times the MIC; the optimal
level would be 10 times the MIC. Similarly, AUIC < 125 times the MIC
indicates suboptimal dosing.
For ß-lactam agents, increasing the duration of time above the MIC
should help prevent the emergence of resistance. For drugs with
relatively short half-lives, levels four to five times the MIC can be
maintained over extended intervals by more frequent dosing or by
continuous infusions.
These predictions are currently theoretical. They are only beginning to
be tested in clinical settings.17
If they are confirmed,
however, it will suggest that clinicians need to consider
pharmacodynamic properties when choosing an antibiotic therapy.
Choosing a drug with the appropriate spectrum of activity will always
remain important in treating bacterial infections. However, choosing
the right dose and dosing interval may also be critical to achieving
optimal clinical responses and preventing the emergence of resistant
pathogens.
|
Appendix 1
|
|---|
Dr. Burgess: The MIC I am quoting here is for the
specific pathogen causing the infection—so it is the actual MIC, not
the MIC50 or MIC90. When considering
pharmacodynamic relationships, people refer to the MIC50 or
the MIC90, but those are for groups of organisms. Some
people have tried to use the MIC50 to define dosing
regimens, but that is not conservative enough since only half of your
Staphylococcus, Escherichia coli, or other pathogenic
isolates are being inhibited. We should be using the MIC90
to be more conservative, although we are probably then overdosing from
an empiric standpoint. Once you get the data back from the clinical
microbiology laboratory, you will have the specific MIC, which you can
then use for determining dosage.
Dr. Yates: I am concerned about increasing the dose of
ciprofloxacin to q8h and its effect on resistance, because we have not
had good results increasing ciprofloxacin dosing.
Dr. Burgess: Increasing the ciprofloxacin dosage from
400 mg q12h to 400 mg q8h has no benefit, because the AUC:MIC ratio is
still inadequate. You still need to add a second drug. For example, a
patient with normal renal function and an infection with a Pseudomonas
species with an MIC of 0.5 µg/mL, which we would call ciprofloxacin
sensitive, would need > 3 g/d—and you cannot give that. You have to
rely on synergy with a second drug. We have always seen synergy with
ß-lactams and aminoglycosides, but it is harder to show synergy with
the quinolones.
Dr. Rapp: We know that with aminoglycosides we can minimize
the effects on human cells by having drug-free intervals. What kind of
side effects are we having on mammalian cells with these other drugs
during continuous infusion?
Dr. Burgess: We really do not know. We do know that with
continuous infusion of ceftazidime, we can use less drug and maintain
the dynamics a lot better—for example, staying at four times the MIC.
But with regard to the adverse reaction profile for continuous infusion
in the clinic, we do not have enough data.
|
Footnotes
|
|---|
Correspondence to: David S. Burgess, PharmD, The University of
Texas Health Science Center, 7703 Floyd Curl Dr, San Antonio, TX
78284-6220; e-mail: burgessd@uthscsa.edu
Abbreviations:
AUC = area under the time-concentration curve; AUIC = area under
the inhibitory curve; MIC = minimal inhibitory concentration;
PAE = postantibiotic effect
|
References
|
|---|
- Craig, WA (1998) Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin Infect Dis 26,1-12[ISI][Medline]
- Craig, WA, Redingtion, J, Ebert, SC (1991) Pharmacodynamics of amikacin in vitro and in mouse thigh and lung infections. J Antimicrob Chemother 27(suppl c),29-40
- Vogelman, B, Gudmundsson, S, Turnidge, J, et al (1988) In vivo postantibiotic effect in a thigh infection in neutropenic mice. J Infect Dis 157,287-298[ISI][Medline]
- McGrath, BJ, Marchbanks, CR, Gilbert, D, et al (1993) In vitro postantibiotic effect following repeated exposure to imipenem, temafloxacin, and tobramycin. Antimicrob Agents Chemother 37,1723-1725[Abstract/Free Full Text]
- den Hollander, JG, Mouton, JW, van Goor, M-LPJ (1996) Alteration of postantibiotic effect during one dosing interval of tobramycin, simulated in an in vitro pharmacokinetic model. Antimicrob Agents Chemother 40,784-786[Abstract]
- Fantin, B, Ebert, S, Leggett, J, et al (1990) Factors affecting duration of in-vivo postantibiotic effect for aminoglycosides against Gram-negative bacilli. J Antimicrob Chemother 27,829-836
- Craig, WA (1993) Post-antibiotic effects in experimental infection models: relationship to in-vitro phenomena and to treatment of infections in man. J Antimicrob Chemother 31(suppl D),149-158
- Moore, RD, Lietman, PS, Smith, CR (1987) Clinical response to aminoglycoside therapy: importance of the ratio of peak concentration to minimal inhibitory concentration. J Infect Dis 155,93-99[Medline]
- Drusano, GL, Johnson, DE, Rosen, M, et al (1993) Pharmacodynamics of a fluoroquinolone antimicrobial agent in a neutropenic rat model of Pseudomonas sepsis. Antimicrob Agents Chemother 37,483-490[Abstract/Free Full Text]
- Forrest, A, Nix, DE, Ballow, CH, et al (1993) Pharmacodynamics of intravenous ciprofloxacin in seriously ill patients. Antimicrob Agents Chemother 37,1073-1081[Abstract/Free Full Text]
- Fass, RJ, Barnishan, J, Solomon, MC, et al (1996) In vitro activities of quinolones, ß-lactams, tobramycin, and trimethoprim-sulfamethoxazole against nonfermentative Gram-negative bacilli. Antimicrob Agents Chemother 40,1412-1418[Abstract]
- Craig, WA (1995) Interrelationship between pharmacokinetics and pharmacodynamics in determining dosage regimens for broad-spectrum cephalosporins. Diagn Microbiol Infect Dis 22,89-96[CrossRef][ISI][Medline]
- Nicolau, DP, Quintiliani, R, Nightingale, CH (1995) Antibiotic kinetics and dynamics for the clinician. Med Clin North Am 79,477-495[ISI][Medline]
- Nicolau, DP, Nightingale, CH, Banevicius, MA, et al (1996) Serum bactericidal activity of ceftazidime: continuous infusion versus intermittent injections. Antimicrob Agents Chemother 40,61-64[Abstract]
- Benko, AS, Cappelletty, DM, Kruse, JA, et al (1996) Continuous infusion versus intermittent administration of ceftazidime in critically ill patients with suspected Gram-negative infections. Antimicrob Agents Chemother 40,691-695[Abstract]
- Schentag, JJ (1998) Antibiotic dosing—does one size fit all? [editorial; comment]. JAMA 279,159-160[Free Full Text]
- Thomas, JK, Forrest, A, Bhavnani, SM, et al (1998) Pharmacodynamic evaluation of factors associated with the development of bacterial resistance in acutely ill patients during therapy. Antimicrob Agents Chemother 42,521-527[Abstract/Free Full Text]
- Craig, WA, Andes, D (1996) Pharmacokinetics and pharmacodynamics of antibiotics in otitis media. Pediatr Infect Dis J 15,255-259[CrossRef][ISI][Medline]
TOP
Abstract
Introduction
