Use of Pharmacokinetics and Pharmacodynamics to Optimize Antimicrobial Treatment of Pseudomonas aeruginosa Infections

SUPPLEMENT ARTICLE Use of Pharmacokinetics and Pharmacodynamics to Optimize Antimicrobial Treatment of Pseudomonas aeruginosa Infections David S. Burgess College of Pharmacy, University of Texas at Austin, Austin, and Departments of Pharmacology and Medicine, University of Texas Health Science Center at San Antonio, San Antonio The study of pharmacodynamics has greatly enhanced our understanding of antimicrobials and has enabled us to optimize dosing regimens. Applying this knowledge to the clinical setting can be critical for the treatment of Pseudomonas aeruginosa infections. Because of its selectively permeable outer membrane and multiple efflux pump mechanisms, P. aeruginosa has high intrinsic resistance to many available antimicrobials. Numerous studies have established pharmacodynamic values for concentration-dependent agents (maximum serum concentration:minimum inhibitory concentration [MIC] and area under the serum concentration time curve: MIC) and concentration-independent agents (i.e., percentage of time that the drug concentration remains greater than the MIC) that help predict the probability of a successful outcome. Current therapies attempt to meet these target values. However, to reduce the risk of clinical failures, combination therapy (typically, a b- lactam with an aminoglycoside or fluoroquinolone) is commonly used to enhance eradication rates and decrease the risk of developing resistance. Although combination therapy ensures a greater chance of selection of appropriate treatment, timely initial administration of antimicrobial therapy remains a key factor for reducing the likelihood of death for these patients. From the time that the first antibiotics were developed, determining the appropriate drug regimen has been challenging and, at times, somewhat controversial. Fortunately, with greater understanding of antimicrobial pharmacodynamics, rational approaches for determining optimal dosing regimens can be pursued to ensure use of the most effective regimen to achieve a successful clinical outcome. Using pharmacokinetic (PK)/pharmacodynamic (PD) relationships to optimize therapy could be critical when treating difficult infections, particularly infections caused by Pseudomonas aeruginosa. This organism possesses a selectively permeable outer membrane, along with several efflux pump mechanisms, that confer intrinsic resistance to multiple antimicrobial classes [1 3]. Antimicrobial treatment of P. Reprints or correspondence: Dr. David S. Burgess, Clinical Pharmacy Programs, MSC 6220, University of Texas Health Science Center, 7703 Floyd Curl Dr., San Antonio, TX 78229-3900 (burgessd@uthscsa.edu). Clinical Infectious Diseases 2005; 40:S99 104 2005 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2005/4004S2-0003$15.00 aeruginosa infections commonly results in the development of resistance. Because of the difficulty and severity of treating these infections, numerous studies have investigated the most effective strategies for treating P. aeruginosa infections. The present article will review the necessary PK/PD requirements for effective treatment of P. aeruginosa infections, along with the therapeutic options commonly used today. PK/PD CONSIDERATIONS The antimicrobial classes commonly used to treat P. aeruginosa infections include the b-lactams (including penicillins, cephalosporins, monobactams, and carbapenems), aminoglycosides, and fluoroquinolones. These agents can be classified according to whether they engage in concentration-independent (or time-dependent) or concentration-dependent killing. By use of a combination of in vitro studies, animal models, and clinical data, PD targets have been determined that correlate with a successful outcome for a particular antimicrobial regimen [4 7]. Optimal PK/PD for P. aeruginosa Infections CID 2005:40 (Suppl 2) S99

Concentration-independent (time-dependent) agents. The bactericidal activity of the b-lactams is concentration independent. For these agents, maximum bacterial killing occurs at concentrations 3 4 times the MIC, with further increases in the drug concentration having little effect [5]. The effectiveness of these agents correlates with the percentage of time that the drug concentration remains greater than the MIC (% t 1 MIC) [5]. Review of the literature reveals that the target value for % t 1 MIC can vary depending on the agent and the particular model used in the study (table 1). An early study by Vogelman et al. [4] showed a direct relationship between the % t 1 MIC and the efficacy of a b-lactam against P. aeruginosa infections. Using a neutropenic mouse model, investigators found that maximum efficacy of ticarcillin against P. aeruginosa required a % t 1 MIC value of nearly 100%. A later study by Craig [5] attempted to correlate the antibacterial activity of the cephalosporins with the % t 1 MIC val- ue by use of a neutropenic murine thigh- and lung-infection model. For gram-negative bacteria, the % t 1 MIC value re- quired for a static effect was 35% 40%. However, the maximum bactericidal effect was achieved when the % t 1 MIC value ap- proached 60% 70%. Although maintaining the b-lactam concentration greater than the MIC correlates with efficacy in vivo, some evidence suggests that simply exceeding the MIC does not maximize the antibacterial activity of these agents. By use of time-kill studies performed with various drug concentration:mic ratios, a concentration of 6.6 times the MIC was determined to provide maximum bactericidal activity of ceftazidime against P. aeruginosa [6]. These results were further supported by a clinical study involving 36 patients with gram-negative infections treated with cefepime [7]. For those patients, microbiological success correlated best with the proportion of the dosing interval that cefepime concentrations exceeded 4.3 times the MIC. To achieve 80% microbiological success, cefepime concentrations would have to be 14.3 times the MIC for 83% of the dosing interval. However, some limitations of that study included the fact that only 20 of the 36 patients had positive culture results, and those patients with a positive culture result received adjunctive aminoglycoside therapy. Since % t 1 MIC is widely accepted as an outcome parameter for b-lactam therapy, various strategies can be used to optimize dosing of these agents to achieve maximum bactericidal effect. Some strategies involve using more-frequent daily doses, administering the dose through continuous infusion, or using concomitant inhibitors of antimicrobial clearance. Several PK studies have shown that continuous infusion will increase the probability of achieving PD targets for the b-lactams [8 10]. However, the results of clinical studies comparing continuous infusion with intermittent dosing have been inconclusive. In fact, a well-designed, randomized, controlled study has yet to be performed to differentiate between the 2 dosing regimens. Clearly, more in vitro and clinical data are needed to determine the optimal % t 1 MIC value for treating P. aeruginosa infections and to distinguish which dosing strategy will optimize outcomes for patients treated with b-lactams. Concentration-dependent agents. Unlike the b-lactams, the PD parameter maximum serum concentration (C max ):MIC is used to assess the effectiveness of concentration-dependent agents. The higher the concentration achieved, the faster the rate of bacterial killing. Typically, a C max :MIC ratio of 8 10 has been associated with effective treatment of and prevention of resistance in gram-negative and -positive bacteria [11, 12]. For this reason, optimal dosing strategies for these agents should attempt to achieve as high a peak value as possible. If the C max :MIC target value cannot be achieved, then the area under the serum concentration time curve (AUC):MIC value may provide an additional parameter with which to judge a drug s effectiveness. The aminoglycosides have been used for 150 years, but only recently has dosing of this antimicrobial class been optimized for the most effective bacterial killing. Several in vitro and in vivo animal studies have shown the concentration-dependent nature of this class of antibiotics for gram-negative infections [11, 13]. Early studies with netilmicin revealed that a C max :MIC ratio of 8 was required to prevent regrowth of P. aeruginosa [11]. In general, a C max :MIC ratio of 8 10 has been accepted as achieving a maximum bactericidal effect against gram-negative pathogens, although few studies have looked specifically at P. aeruginosa (table 2) [12]. The concentration-dependent nature of the aminoglycosides has been observed in animal models [4, 13]. In a neutropenic murine infection model, the AUC showed the greatest corre- Table 1. b-lactam pharmacodynamic (PD) target values for effective treatment of Pseudomonas aeruginosa infection. Antimicrobial agent [reference] PD target value Source model Ticarcillin [4] % t 1 MIC, 100% Cefotaxime, ceftriaxone, ceftazidime, and cefpirome [5] % t 1 MIC, 35%; static % t 1 MIC, 60% 70% (to prevent regrowth) Neutropenic mouse thigh infection Neutropenic mouse thigh and lung infections Ceftazidime [6] Maintain concentration 6.6 the MIC In vitro time-kill analysis Cefepime [7] % t 1 MIC, 14.3 the MIC (83% 95%) NOTE. % t 1 MIC, percentage of time the drug concentration remains greater than the MIC. Human S100 CID 2005:40 (Suppl 2) Burgess

Table 2. Aminoglycoside pharmacodynamic (PD) target values for effective treatment of Pseudomonas aeruginosa infection. Antimicrobial agent [reference] PD target value Source model Netilmicin [11] C max :MIC1 8 Neutropenic mouse thigh infection Tobramycin [4] AUC 1 30 Neutropenic mouse thigh infection Gentamicin, tobramycin, and amikacin [14] C max 6 mg/ml for gentamicin and tobramycin; C max 24 mg/ml for amikacin Human, gram-negative infections Gentamicin, tobramycin, and amikacin [15] C max :MIC 8 Human, gram-negative infections NOTE. AUC, area under the serum concentration time curve; C max, maximum serum concentration. lation with efficacy for treating gram-negative infections. Evidence from early clinical studies showed agreement with the results from in vitro and animal models, establishing a relationship between serum levels of aminoglycoside and clinical efficacy [14 16]. One study was able to correlate efficacy with the ratio of the serum concentration of aminoglycoside to the MIC of the pathogen [15]. On the basis of these data, improved dosing regimens were investigated to achieve optimal clinical efficacy with the aminoglycosides [17 19]. Traditional dosing of the aminoglycosides typically involved administering a low dose of the agent 2 3 times/day [17]. With the increasing PD evidence of optimal aminoglycoside target values, new dosing strategies were explored to maximize the concentration-dependent nature of these antimicrobials. Oncedaily dosing that uses higher amounts of the drug has been explored and was found to achieve efficacy that was at least comparable to that of traditional dosing regimens while reducing the risk of nephrotoxicity [18, 19]. Once-daily dosing can also increase the cost-effectiveness of aminoglycoside therapy by reducing preparation and administration time. An extended dosing interval is possible with the use of aminoglycosides because of the postantibiotic effect (PAE) of these agents. The PAE is the suppression of bacterial growth after limited exposure of the organisms to the antibiotic. A number of factors can influence the duration of the PAE, including the organism and its MIC, the duration of exposure, the concentration of the drug, and the potency of the agent. The b-lactams do not show a significant PAE for gram-negative organisms (with the exception of carbapenems), although they do exhibit a PAE for gram-positive organisms. For the aminoglycosides, studies have shown that, the higher the concentration of drug, the longer the duration of the PAE. The fluoroquinolones were the first class of drugs to use extensive PD data to develop dosing regimens. Similar to the findings for aminoglycosides, in vitro models showed that a C max :MIC ratio of 10 was required for effective killing and prevention of regrowth of resistant subpopulations. A second parameter was the AUC:MIC ratio, for which a value of 100 125 was recommended for gram-negative infections (although a ratio of only 30 40 is required for gram-positive infections) [20]. These results were further supported by animal studies and clinical data (table 3). Using in vitro kill-curve analysis, MacGowan et al. [23] demonstrated that levofloxacin and ciprofloxacin achieved similar bactericidal effects against 3 strains of P. aeruginosa once the AUC:MIC ratio reached 125. Although ciprofloxacin tends to have MIC values 2-fold lower than those of levofloxacin for P. aeruginosa, levofloxacin achieves plasma and tissue concentrations 2 fold higher than those of ciprofloxacin [20]. Both agents should then provide similar activity against these infections in a patient. Ciprofloxacin and levofloxacin are the only fluoroquinolones commercially available that are approved by the Food and Drug Administration for the treatment of systemic P. aeruginosa infections [24, 25]. As with the aminoglycosides, optimal treatment of P. aeruginosa infections with fluoroquinolones would require achieving as high a concentration as clinically possible Table 3. Fluoroquinolone pharmacodynamic (PD) target values for effective treatment of Pseudomonas aeruginosa infection. Antimicrobial agent [reference] PD target value Source model Enoxacin [11] C max :MIC 18 IVPM Ciprofloxacin and ofloxacin [42] AUC:MIC 1100 IVPM Lomefloxacin [21] C max :MIC 110 Neutropenic rat model Ciprofloxacin [22] AUC:MIC 1100 Human, retrospective Levofloxacin and ciprofloxacin [30] AUC:MIC 1157 Mouse, mathematical modeling Levofloxacin and ciprofloxacin [23] AUC:MIC 125 In vitro time-kill curves NOTE. AUC, area under the serum concentration time curve; C max, maximum serum concentration; IVPM, in vitro pharmacokinetic model. Optimal PK/PD for P. aeruginosa Infections CID 2005:40 (Suppl 2) S101

to attain the PD target value for a successful clinical outcome. A high-dose regimen of each of these drugs has been approved for severe infections. The high-dose ciprofloxacin regimen increases the frequency of dosing (400 mg iv t.i.d., compared with the traditional regimen of 400 mg iv b.i.d.), which increases the AUC from 12.7 m g h/ml to 32.9 m g h/ml, but does not increase the C max (4.56 vs. 4.07 mg/ml for the b.i.d. vs. t.i.d. dosing regimens, respectively) [24]. Higher doses (1400 mg) of ciprofloxacin are not recommended, because a potential increased incidence of adverse events involving the CNS has been associated with increased plasma concentrations of this agent [26, 27]. For levofloxacin, a high-dose regimen of 750 mg q.d. results in approximately double the AUC (108 mg h/ml vs. 54.6 mg h/ml) and C max (12.1 mg/ml vs. 6.4 mg/ ml) values, compared with those for the 500 mg q.d. dose [25]. The higher C max and AUC values increase the probability of attaining the necessary PD target values for successful clinical outcomes [28, 29]. Fortunately, the safety profile of the 750- mg levofloxacin dose is similar to that of the lower dose, allowing maximum drug exposure with once-daily dosing [25]. A recent study used a mathematical model to predict the probability of clinical success associated with the use of fluoroquinolone treatment for P. aeruginosa infections [30]. By use of population pharmacodynamics, an AUC:MIC ratio of 157 was determined to be the target to limit amplification of resistant subpopulations of P. aeruginosa during an infection. When Monte Carlo simulation and the MIC distribution of P. aeruginosa from a surveillance study were used, the probability that a 750-mg q.d. levofloxacin regimen would meet this target was 61.2%, compared with 61.8% for a ciprofloxacin regimen of 400 mg t.i.d. Although no difference exists between levofloxacin and ciprofloxacin in achieving this desired target of an AUC:MIC ratio of 157, neither agent would be considered adequate for use as monotherapy against this organism. Other available fluoroquinolones would not achieve a higher rate because of higher MIC values and/or lower drug exposure, compared with levofloxacin and ciprofloxacin. Finally, additional consideration must be given when choosing antimicrobial doses for seriously ill patients because of their altered pharmacokinetics. These patients tend to have differences in drug clearance, volume distribution, and elimination half-life than do healthy individuals, even when renal and hepatic functions appear to be normal. Rebuck et al. [31] showed significant increases in C max and AUC values of critically ill patients treated with levofloxacin, compared with those of healthy volunteers. In general, drug exposures are higher in critically ill and elderly patients, and they may improve the probability of attaining necessary PK/PD targets for effective antimicrobial therapy [32]. However, careful monitoring and dose adjustment may be necessary to minimize the risk of adverse events. COMBINATION THERAPY Recommended treatments for systemic P. aeruginosa infections typically involve combination therapy, since it increases the chance of effective initial therapy before receiving susceptibility results. Combination therapy may minimize the risk of developing resistance and has the potential for synergistic activity. For serious P. aeruginosa infections, combination therapy has traditionally included an antipseudomonal b-lactam (such as piperacillin, ceftazidime, cefepime, aztreonam, imipenem, or meropenem) with an aminoglycoside or a fluoroquinolone. Aminoglycosides have been shown to confer in vitro synergy when combined with a b-lactam. However, the aminoglycosides are also associated with a greater risk of adverse events, such as nephrotoxicity and ototoxicity. The fluoroquinolones also display synergy when combined with a b-lactam, although usually not to such a high degree as the aminoglycoside/b-lactam combinations [33, 34]. In addition, the fluoroquinolones are available in iv and oral doses, allowing for convenient switching of therapy during the therapeutic course. In vitro studies evaluating antimicrobial combinations. In vitro synergy studies have produced wide ranges of results regarding the frequency of synergy with b-lactam combinations involving fluoroquinolones or aminoglycosides. The synergy results are dependent on several methodological factors, such as the method of synergy testing used (i.e., checkerboard or time-kill analysis), the susceptibility patterns of the isolates, and the concentrations of antimicrobials used. In general, time-kill analysis demonstrates a higher degree of synergy than does checkerboard analysis and has been better correlated with clinical outcome for P. aeruginosa infection [35]. Although standardized definitions of synergy exist for either checkerboard or time-kill analyses, standardization for each analysis does not exist. Earlier work from our laboratory using time-kill analysis showed that b-lactams combined with gentamicin resulted in synergy at a frequency of 79% when tested against 12 P. aeruginosa isolates [34]. Although this frequency was higher Figure 1. Comparison of killing rates of gentamicin ( ), levofloxacin ( ), and ciprofloxacin ( ) against Pseudomonas aeruginosa strain 99-036., Growth control. S102 CID 2005:40 (Suppl 2) Burgess

than that associated with b-lactams combined with ciprofloxacin (58%) or levofloxacin (67%), the difference was not significant. Furthermore, the weaker the antimicrobial agent is against P. aeruginosa, the more likely and easier it becomes to demonstrate synergy. For example, as displayed in figure 1, the in vitro killing by gentamicin alone is less than that by levofloxacin alone, which is less than that by ciprofloxacin alone. Therefore, on the basis of the standard definition of synergy that a 2-log 10 reduction is needed with the combination, compared with the most active agent alone at 24 h, one can clearly see that, for this isolate of P. aeruginosa, synergy would be more easily observed for gentamicin, followed by levofloxacin and then ciprofloxacin. Hence, synergy may not provide the most practical means for comparing the effectiveness of all antimicrobial combinations, particularly if one or both agents in a combination exhibit a high degree of potency against a particular pathogen (and, thus, would have a low likelihood of measurable synergy). Time-kill analyses evaluating bactericidal activity may provide an additional indication of how effectively certain combinations can eradicate a pathogen. One study compared the bactericidal activity of piperacillin-tazobactam used in combination with either gentamicin or ciprofloxacin for 6 strains of P. aeruginosa [36]. Although both combinations showed synergy for all 6 strains, the ciprofloxacin combination resulted in a greater amount of bacterial killing after 24 h. In a study from our laboratory comparing the bactericidal activity of levofloxacin, ciprofloxacin, and gentamicin used in combination with piperacillin-tazobactam and cefepime against 12 strains of P. aeruginosa [34], we found no difference between levofloxacin and ciprofloxacin combinations in bactericidal activity at 24 h. In fact, the fluoroquinolones were bactericidal at 24 h for 90% of the combinations, whereas the aminoglycosides were bactericidal for 100% of the combinations. The fluoroquinolone combinations maintained bactericidal killing for all but 2 isolates, which were nonsusceptible to the fluoroquinolones and had an MIC 64 mg/ml for piperacillin-tazobactam [34]. In another study, we compared the bactericidal activity of 500 and 750 mg of levofloxacin in combination with ceftazidime, cefepime, piperacillin-tazobactam, imipenem, and tobramycin against 12 strains of P. aeruginosa [37]. The bactericidal activity was maintained more often with 750 mg than with 500 mg of levofloxacin (94% vs. 83%). In fact, the bactericidal activity was related to the susceptibility of the isolates to levofloxacin. For those isolates that were susceptible to levofloxacin, no difference in bactericidal activity was noted for the 500- and 750-mg levofloxacin regimens. However, for levofloxacin-intermediate and -resistant isolates, the bactericidal activity was different for the 2 dosing regimens, with the 750-mg regimen providing greater bactericidal activity (for intermediate isolates, synergy was observed in 75% and 100% of combinations with 500 and 750 mg of levofloxacin, respectively; for resistant isolates, the rates were 38% and 63% for combinations with 500 and 750 mg of levofloxacin, respectively) [37]. In a report included in this supplement, Lister [38] used an in vitro PK model to show that a levofloxacin-imipenem combination effectively eradicated 3 clinical isolates of P. aeruginosa, whereas monotherapy with either agent allowed regrowth and the development of resistance. Because of the methodology issues of synergy testing as discussed, it remains difficult to determine whether in vitro results correlated with clinical efficacy. To date, no large-scale, prospective randomized clinical trials have specifically compared monotherapy with combination therapy or have compared different combination therapies for treating P. aeruginosa infections. Information has predominantly relied on retrospective analysis or extraction of data from comparative clinical trials, usually for the treatment of nosocomial pneumonia. Studies have shown that the use of combination therapy ensures a greater likelihood of selection of initial adequate treatment for P. aeruginosa infections and results in significantly decreased mortality [39 41]. For P. aeruginosa infections, adequate initial combination therapy administered in a timely manner must be emphasized to optimize the chance of clinical success and decrease the risk of emergence of resistance. SUMMARY Systemic P. aeruginosa infections remain among the most challenging conditions facing physicians, because of the organism s high intrinsic resistance and its ability to acquire adaptive resistance during a course of therapy. However, the probability of a successful clinical outcome can be improved with a thorough understanding of the antibiotic PK/PD profile against this organism. As the field of pharmacodynamics has expanded over the years, dosing regimens have been optimized in vitro to capitalize on the concentration-dependent or -independent nature of antimicrobials, ensuring maximum bacterial killing and reducing the risk of developing resistance. Certain combinations of antimicrobials provide synergistic activity against P. aeruginosa, and, in limited studies, combination therapy increases the likelihood of selection of initial adequate therapy for patients. Providing timely and adequate antimicrobial therapy has proven to be critical to achieving successful clinical outcomes for these difficult-to-treat infections. Acknowledgments Financial support. Abbott; AstraZeneca; Merck; Ortho-McNeil; Wyeth. Potential conflicts of interest. D.S.B. is a consultant to Abbott, Astra- Zeneca, Aventis, Merck, Ortho-McNeil, and Wyeth and serves on the speakers bureaus of Abbott, AstraZeneca, Aventis, Merck, Ortho-McNeil, and Wyeth. Optimal PK/PD for P. aeruginosa Infections CID 2005:40 (Suppl 2) S103

References 1. Hancock RE. Resistance mechanisms in Pseudomonas aeruginosa and other nonfermentative gram-negative bacteria. Clin Infect Dis 1998; 27(Suppl 1):S93 9. 2. Nikaido H. Antibiotic resistance caused by gram-negative multidrug efflux pumps. Clin Infect Dis 1998; 27(Suppl 1):S32 41. 3. Aeschlimann JR. The role of multidrug efflux pumps in the antibiotic resistance of Pseudomonas aeruginosa and other Gram-negative bacteria. Pharmacotherapy 2003; 23:916 24. 4. Vogelman B, Gudmundsson S, Leggett J, Turnidge J, Ebert S, Craig WA. Correlation of antimicrobial pharmacokinetic parameters with therapeutic efficacy in an animal model. J Infect Dis 1988; 158:831 47. 5. Craig WA. Interrelationship between pharmacokinetics and pharmacodynamics in determining dosage regimens for broad-spectrum cephalosporins. Diagn Microbiol Infect Dis 1995; 22:89 96. 6. Manduru M, Mihm LB, White RL, Friedrich LV, Flume PA, Bosso JA. In vitro pharmacodynamics of ceftazidime against Pseudomonas aeruginosa isolates from cystic fibrosis patients. Antimicrob Agents Chemother 1997; 41:2053 6. 7. Tam VH, McKinnon PS, Akins RL, Rybak MJ, Drusano GL. Pharmacodynamics of cefepime in patients with Gram-negative infections. J Antimicrob Chemother 2002; 50:425 8. 8. Turnidge JD. The pharmacodynamics of b-lactams. Clin Infect Dis 1998; 27:10 22. 9. Burgess DS, Hastings RW, Hardin TC. Pharmacokinetics and pharmacodynamics of cefepime administered by intermittent and continuous infusion. Clin Ther 2000; 22:66 75. 10. Tam VH, Louie A, Lomaestro BM, Drusano GL. Integration of population pharmacokinetics, a pharmacodynamic target, and microbiologic surveillance data to generate a rational empiric dosing strategy for cefepime against Pseudomonas aeruginosa. Pharmacotherapy 2003; 23:291 5. 11. Blaser J, Stone BB, Groner MC, Zinner SH. Comparative study with enoxacin and netilmicin in a pharmacodynamic model to determine importance of ratio of antibiotic peak concentration to MIC for bactericidal activity and emergence of resistance. Antimicrob Agents Chemother 1987; 31:1054 60. 12. Lacy MK, Nicolau DP, Nightingale CH, Quintiliani R. The pharmacodynamics of aminoglycosides. Clin Infect Dis 1998; 27:23 7. 13. Leggett JE, Fantin B, Ebert S, et al. Comparative antibiotic dose-effect relations at several dosing intervals in murine pneumonitis and thighinfection models. J Infect Dis 1989; 159:281 92. 14. Moore RD, Smith CR, Lietman PS. Association of aminoglycoside plasma levels with therapeutic outcome in gram-negative pneumonia. Am J Med 1984; 77:657 62. 15. Moore RD, Lietman PS, Smith CR. Clinical response to aminoglycoside therapy: importance of the ratio of peak concentration to minimal inhibitory concentration. J Infect Dis 1987; 155:93 9. 16. Moore RD, Smith CR, Lietman PS. The association of aminoglycoside plasma levels with mortality in patients with gram-negative bacteremia. J Infect Dis 1984; 149:443 8. 17. Kashuba AD, Nafziger AN, Drusano GL, Bertino JS. Optimizing aminoglycoside therapy for nosocomial pneumonia caused by gram-negative bacteria. Antimicrob Agents Chemother 1999; 43:623 9. 18. Nicolau DP, Freeman CD, Belliveau PP, Nightingale CH, Ross JW, Quintiliani R. Experience with a once-daily aminoglycoside program administered to 2,184 adult patients. Antimicrob Agents Chemother 1995; 39:650 5. 19. Murry KR, McKinnon PS, Mitrzyk B, Rybak MJ. Pharmacodynamic characterization of nephrotoxicity associated with once-daily aminoglycoside. Pharmacotherapy 1999; 19:1252 60. 20. Andes D, Craig WA. Animal model pharmacokinetics and pharmacodynamics: a critical review. Int J Antimicrob Agents 2002; 19:261 8. 21. Drusano GL, Johnson DE, Rosen M, Standiford HC. Pharmacodynamics of a fluoroquinolone antimicrobial agent in a neutropenic rat model of Pseudomonas sepsis. Antimicrob Agents Chemother 1993; 37:483 90. 22. Thomas JK, Forrest A, Bhavnani SM, et al. Pharmacodynamic evaluation of factors associated with the development of bacterial resistance in acutely ill patients during therapy. Antimicrob Agents Chemother 1998; 42:521 7. 23. MacGowan AP, Wootton M, Holt HA. The antibacterial efficacy of levofloxacin and ciprofloxacin against Pseudomonas aeruginosa assessed by combining antibiotic exposure and bacterial susceptibility. J Antimicrob Chemother 1999; 43:345 9. 24. Bayer Pharmaceuticals. Cipro (ciprofloxacin) I.V. [package insert]. West Haven, CT: Bayer Pharmaceuticals, 2002. 25. Ortho-McNeil Pharmaceutical. Levaquin tablets/injection [package insert]. Raritan, NJ: Ortho-McNeil Pharmaceutical, 2002. 26. Echols RM. The approval process for intravenous ciprofloxacin. Infect Med 1992; 9:23 31. 27. Echols RM. The selection of appropriate dosages for intravenous ciprofloxacin. J Antimicrob Chemother 1993; 31:783 7. 28. Gotfried MH, Danziger LH, Rodvold KA. Steady-state plasma and intrapulmonary concentrations of levofloxacin and ciprofloxacin in healthy adult subjects. Chest 2001; 119:1114 22. 29. Rodvold KA, Danziger LH, Gotfried MH. Steady-state plasma and bronchopulmonary concentrations of intravenous levofloxacin and azithromycin in healthy adults. Antimicrob Agents Chemother 2003; 47:2450 7. 30. Jumbe N, Louie A, Leary R, et al. Application of a mathematical model to prevent in vivo amplification of antibiotic-resistant bacterial populations during therapy. J Clin Invest 2003; 112:275 85. 31. Rebuck JA, Fish DN, Abraham E. Pharmacokinetics of intravenous and oral levofloxacin in critically ill adults in a medical intensive care unit. Pharmacotherapy 2002; 22:1216 25. 32. Goldberg J, Owens RC Jr. Optimizing antimicrobial dosing in the critically ill patient. Curr Opin Crit Care 2002; 8:435 40. 33. Burgess DS, Hastings RW. Activity of piperacillin/tazobactam in combination with amikacin, ciprofloxacin, and trovafloxacin against Pseudomonas aeruginosa by time-kill. Diagn Microbiol Infect Dis 2000; 38: 37 41. 34. Burgess DS, Nathisuwan S. Cefepime, pipericillin/tazobactam, gentamicin, ciprofloxacin, and levofloxacin alone and in combination against Pseudomonas aeruginosa. Diagn Microbiol Infect Dis 2002; 44:35 41. 35. Hilf M, Yu VL, Sharp J, Zuravleff JJ, Korvick JA, Muder RR. Antibiotic therapy for Pseudomonas aeruginosa bacteremia: outcome correlations in a prospective study of 200 patients. Am J Med 1989; 87:540 6. 36. Gould IM, Milne K. In-vitro pharmacodynamic studies of piperacillin/ tazobactam with gentamicin and ciprofloxacin. J Antimicrob Chemother 1997; 39:53 61. 37. Burgess DS, Hall RG, Hardin TC. In vitro evaluation of the activity of two doses of levofloxacin alone and in combination with other agents against Pseudomonas aeruginosa. Diagn Microbiol Infect Dis 2003; 46: 131 7. 38. Lister PD, Wolter DJ. Levofloxacin-imipenem combination prevents the emergence of resistance among clinical isolates of Pseudomonas aeruginosa. Clin Infect Dis 2005; 40(Suppl 2):S105 14 (in this issue). 39. Luna CM, Vujacich P, Niederman MS, et al. Impact of BAL data on the therapy and outcome of ventilator-associated pneumonia. Chest1997; 111:676 85. 40. Kollef MH, Sherman G, Ward S, Fraser VJ. Inadequate antimicrobial treatment of infections: a risk factor for hospital mortality among critically ill patients. Chest 1999; 115:462 74. 41. Ibrahim EH, Ward S, Sherman G, Schaiff R, Fraser VJ, Kollef MH. Experience with a clinical guideline for the treatment of ventilatorassociated pneumonia. Crit Care Med 2001; 29:1109 15. 42. Madaras-Kelly KJ, Larsson AJ, Rotschafer JC. A pharmacodynamic evaluation of ciprofloxacin and ofloxacin against two strains of Pseudomonas aeruginosa. J Antimicrob Chemother 1996; 37:703 10. S104 CID 2005:40 (Suppl 2) Burgess