Pharmacodynamics of doripenem in critically ill patients with ventilator-associated Gram-negative bacilli pneumonia

International Journal of Antimicrobial Agents 40 (2012) 434–439

Contents lists available at SciVerse ScienceDirect

International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag

Pharmacodynamics of doripenem in critically ill patients with ventilator-associated Gram-negative bacilli pneumonia Sutep Jaruratanasirikul a,∗ , Wibul Wongpoowarak b , Narongdet Kositpantawong a , Nanchanit Aeinlang a , Monchana Jullangkoon a a b

Department of Medicine, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkla 90110, Thailand Department of Pharmaceutical Technology, Faculty of Pharmaceutical Sciences, Prince of Songkla University, Hat Yai, Songkla 90110, Thailand

a r t i c l e

i n f o

Article history: Received 23 April 2012 Accepted 18 July 2012 Keywords: Pharmacokinetics/pharmacodynamics Pharmacodynamics Population pharmacokinetics Doripenem Carbapenem Ventilator-associated pneumonia

a b s t r a c t Several pathophysiological changes in critically ill patients are important in determining the therapeutic success of ␤-lactam antibiotics. The aim of this study was to assess the population pharmacokinetics and probabilities of target attainment (PTAs) of doripenem in patients with ventilator-associated pneumonia, comparing administration by 1-h and 4-h infusion. Patients were randomised into two groups: Group I received a 1-h infusion of 0.5 g every 8 h (q8h) for seven doses; and Group II received a 4-h infusion of 0.5 g q8h for seven doses. A Monte Carlo simulation was performed to determine the PTAs. PTAs of achieving 40% T>MIC [exposure time during which the free drug concentration remains above the minimum inhibitory concentration (MIC)] and 75% T>MIC are required for effective bactericidal activity of this agent in immunocompetent and immunocompromised hosts, respectively. Values of volume of distribution and total clearance of doripenem in these patients were 17.26 ± 1.83 L and 24.89 ± 1.63 L/h, respectively. For pathogens with a MIC of 1 ␮g/mL, the PTAs of achieving 40% T>MIC following administration of doripenem by a 1-h and 4-h infusion of 0.5 g q8h were 92.95% and 98.32%, respectively. For pathogens with a MIC of 2 ␮g/mL in immunocompromised hosts, the PTAs of achieving 80% T>MIC following administration of doripenem by 1-h and 4-h infusion of 2 g q8h were 56.57% and 91.21%, respectively. In conclusion, these findings indicated that higher than recommended doses in this patient population, particularly neutropenic patients, would be necessary to optimise the pharmacokinetics of doripenem. © 2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.

1. Introduction In the current era of increasing multidrug-resistant (MDR) bacteria in hospitals, especially Pseudomonas aeruginosa and Acinetobacter spp., as well as only a small number of new and effective antimicrobial agents, treating infections caused by these microorganisms is becoming more difficult [1,2]. One important strategy to deal with this problem is optimising the antimicrobial activity of current antibiotics by finding the optimum dosing strategy. Among the classes of antimicrobial agents widely prescribed in nosocomial infections, ␤-lactams are the most important antibiotics used for coverage of these highly resistant pathogens. Doripenem, a carbapenem antibiotic, is a ␤-lactam antimicrobial agent with a broad spectrum of activity. It is approved for the treatment of complicated urinary tract infections (UTIs) and complicated intra-abdominal infections in the USA and Europe

∗ Corresponding author. Tel.: +66 74 451 452; fax: +66 74 429 385. E-mail address: [email protected] (S. Jaruratanasirikul).

and is also approved for nosocomial pneumonia in Europe [3]. This agent, in common with other ␤-lactams, is characterised by time-dependent antimicrobial activity, and the exposure time during which the free drug concentrations remains above the minimum inhibitory concentration (T>MIC ) is the pharmacokinetic/pharmacodynamic (PK/PD) index that best correlates with efficacy [4,5]. The mortality rate of patients with nosocomial pneumonia, including ventilator-associated pneumonia (VAP), remains high [6]. Inappropriate use of antibiotics in terms of spectrum of antimicrobial activity, dosage regimens, frequency of administration and duration of therapy can cause therapeutic failure or delayed response in critically ill patients. In addition, PK variations, including increased volume of distribution (Vd ) and drug clearance, of hydrophilic antimicrobial agents in this patient population may occur compared with healthy subjects, resulting in fluctuations of plasma concentrations and the PK/PD index [7,8]. The current study is the first pharmacodynamic study of doripenem in critically ill patients with VAP. A Monte Carlo simulation (MCS) was also used to forecast the efficacy of dosage regimens of doripenem. The aim of the study was to describe the population pharmacokinetics and

0924-8579/$ – see front matter © 2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. http://dx.doi.org/10.1016/j.ijantimicag.2012.07.014

S. Jaruratanasirikul et al. / International Journal of Antimicrobial Agents 40 (2012) 434–439

the probability of target attainment (PTA) of doripenem in patients with VAP, comparing 1-h and 4-h infusion administrations. 2. Methods 2.1. Subjects This study was conducted in patients with VAP admitted into the Intensive Care Unit (ICU) of Songklanagarind Hospital, the largest tertiary-care centre in southern Thailand, from October 2010 through October 2011. Patients were eligible for the study if they met the following criteria: (i) >18 years of age; (ii) intubated and receiving mechanical ventilation for ≥48 h; and (iii) clinical suspicion of VAP, defined by a new and persistent infiltrate on chest radiography associated with at least one of the following: purulent tracheal secretions; temperature of ≥38.3 ◦ C; or a leucocyte count >10 000 cells/mm3 . Patients were excluded from the study if they were pregnant or in circulatory shock (defined as a systolic blood pressure of 90 mmHg and poor tissue perfusion) or had documented hypersensitivity to carbapenems or an estimated creatinine clearance (CLCr ) (determined by the Cockcroft–Gault method) [9] of <50 mL/min. The severity of illness of each patient was assessed at the time of enrolment into the study using Acute Physiology and Chronic Health Evaluation (APACHE) II scores and the Sepsis-related Organ Failure Assessment (SOFA) score. Diagnosis of VAP was also evaluated by the Clinical Pulmonary Infection Score (CPIS) [10]. The protocol for the study was approved by the Ethics Committee of Songklanagarind Hospital. Written informed consent was obtained from each subject’s legally acceptable representative before enrolment. 2.2. Drugs and chemicals Doripenem (DoribaxTM ) was generously donated by JanssenCilag Ltd. (Bangkok, Thailand). Doripenem standard powder was generously donated by Johnson & Johnson Pharmaceutical Research & Development (Raritan, NJ). The pharmaceutical companies were not involved with, and did not interfere with, data collection, interpretation of findings or preparation of the manuscript. All the solvents were of high-performance liquid chromatography (HPLC) grade. 2.3. Study design Following the manufacturer’s instructions, the dosage recommendation of doripenem is either 1-h or 4-h infusion of 0.5 g every 8 h (q8h). Patients were allocated by simple randomisation into two groups using blinded opaque envelope selection. Group I received a 1-h infusion of 0.5 g of doripenem diluted in 100 mL of normal saline solution via an infusion pump at a constant flow rate q8h for seven doses, and Group II received a 4-h infusion of 0.5 g of doripenem diluted in 100 mL of normal saline solution via an infusion pump at a constant flow rate q8h for seven doses. A MCS was performed to predict the efficacy of standard dosage regimens (0.5 g q8h) and higher dosage regimens (1 g q8h and 2 g q8h) of doripenem. Each patient received doripenem at room temperature (32–37 ◦ C). After completion of the 3-day study regimen, all patients were appropriately treated with other antibiotics for 10 days. 2.4. Blood sampling Doripenem PK studies were carried out during administration of the seventh dose of each regimen. Blood samples (ca. 3 mL) were obtained by direct venipuncture at the following times: shortly before (time 0) and then at 0.5, 1, 1.5, 2, 4, 5, 6, 7 and 8 h after

435

the seventh dose of a 1-h infusion and shortly before (time 0) and then at 0.5, 1, 2, 3, 4, 4.5, 5, 6, 7 and 8 h after the seventh dose of a 4-h infusion. All blood samples were added to a heparinised tube, immediately stored on ice, and centrifuged at 1000 × g at 4 ◦ C for 10 min within 5 min. An equal volume of stabilising solution [0.5 M 3-(N-morpholino)propanesulfonic acid (MOPS)/water/ethylene glycol, 2:1:1, v/v/v] was added to each plasma sample and the mixture was then vortexed and stored at −80 ◦ C until analysis within 1 week. 2.5. Doripenem assay Concentrations of doripenem were determined by reversedphase HPLC. The samples were prepared by the method of Ikeda et al. [11]. Briefly, 200 ␮L of stabilising solution was added to 200 ␮L of sample. The mixture was then subjected to a centrifugal ultrafiltration device using a Nanosep® 10K (Pall Corporation, Northborough, MA). The devices were centrifuged at 12 000 × g for 20 min at room temperature. A 20 ␮L aliquot of the sample was injected onto a Symmetry C18 column (Waters Associates, Milford, MA) using an automated injection system (Waters 717 Plus Autosampler; Waters Associates). The mobile phase was a mixture of 0.1 M sodium phosphate buffer (pH 4.6) and acetonitrile (95:5, v/v) at a flow rate of 1 mL/min. The column effluent was monitored by photodiode array detection (Waters 2996; Waters Associates) at 300 nm. Peaks were recorded and integrated using Empower (Waters Associates). The limit of detection of doripenem was 0.08 ␮g/mL and the limit of quantitation was 0.27 ␮g/mL. The intra-assay reproducibility values characterised by coefficients of variation (CVs) were 5.67%, 1.89% and 4.75% for samples containing 0.5, 25 and 50 ␮g/mL, respectively. The interassay reproducibility precision values, calculated by CVs, were 7.52%, 7.06% and 7.42% for samples containing 0.5, 25 and 50 ␮g/mL, respectively. 2.6. Pharmacokinetic analysis The differential equation describing the two-compartment infusion model used in the study was solved numerically by using the Taylor series expansion method [12]. Summation of infinite series was performed until convergence was achieved numerically using the Visual Basic programming language in a Microsoft Excel spreadsheet (Microsoft Corp., Redmond, WA). PK parameters were obtained by non-linear regression to minimise the objective function of the sum square errors (SSEs): obj = min



N i=1

 (ln yactual − ln ycalc )2i

where yactual and ycalc were the observed and calculated concentrations, respectively, for all N data points. Since a PK/PD simulation aims at determining the time above the MIC, the elimination-phase behaviour is critical for a reliable prediction near the MIC. With this calculation, a logarithmic scale is used to describe observed concentrations instead of a normal scale to avoid the effect from a large residual error of predictions at a higher concentration zone during the infusion period. The algorithm used for minimisation of the SSEs in this study was random heuristic optimisation as described elsewhere [13,14]. This method has good convergence speed and can be conveniently implemented in a spreadsheet. 2.7. Pharmacodynamic assessment using Monte Carlo simulation The values of PK parameters obtained were not normally distributed and their behaviour could be represented much more

436

S. Jaruratanasirikul et al. / International Journal of Antimicrobial Agents 40 (2012) 434–439

Table 1 Comparison of actual and simulated pharmacokinetic (PK) parameters of doripenem in 11 patients with ventilator-associated pneumonia following administration by 1-h and 4-h infusion. Parameter

Geometric mean

Actual PK parameter 5.309 k12 (/h) 2.639 k21 (/h) 1.442 ke (/h) 24.893 CL (L/h) 17.263 Vd (L) Vd (L/kg) 0.303 Simulated PK parameter 5.302 k12 (/h) 2.613 k21 (/h) 1.451 ke (/h) 24.923 CL (L/h) Vd (L) 17.179

Geometric S.D.

Median (95% CI)

1.941 2.127 1.346 1.637 1.830 1.663

5.069 (1.447–11.545) 2.835 (0.838–7.779) 1.369 (1.009–2.321) 24.762 (9.485–44.032) 21.170 (5.437–38.101) 0.312 (0.121–0.607)

1.954 2.138 1.306 1.703 1.821

5.314 (1.430–19.677) 2.611 (0.598–11.656) 1.445 (0.862–2.452) 24.905 (8.631–70.991) 17.141 (5.325–56.222)

S.D., standard deviation; CI, confidence interval; k12 , intercompartmental transfer rate constant from compartment X1 to X2 ; k21 , intercompartmental transfer rate constant from compartment X2 to X1 ; ke , elimination rate constant from X1 ; CL, total clearance; Vd , volume of distribution.

usefully using a logarithmic scale (Table 1). The MCS used in this study was based on log-normal distribution of PK parameters. The simulation was performed using the Visual Basic language program. A set of PK parameters was simulated from each parameter’s geometric mean and geometric standard deviation (S.D.). The algorithm for a normal distribution generator was a Box–Muller transform [15]. From the PK parameters obtained in this study (Table 1), a MCS was performed to achieve concentration–time profiles. The PK parameters were simulated according to the statistical behaviour of the actual PK parameters and were selected to ensure that their covariances were comparable with those obtained from the actual parameters. Their average, S.D. and covariance matrix of simulated

parameters were very close to the actual values as shown in Table 1 and Appendix A. The simulated parameters were used in solving the differential equations describing the two-compartment model used in the study. By comparing with the given MIC, percentage time above the MIC (%T>MIC ) could be computed and recorded. We used 10 000 simulations to see the behaviour of 40% T>MIC and 80% T>MIC . 2.8. Validation of suitability The simulated PK parameters were compared with the parameters obtained from non-linear regression. They were found to be comparable with each other (Table 1). 3. Results Eleven patients were enrolled in the study (ten males and one female). Their mean age was 50 ± 16 years (range 25–80 years) and their mean weight was 57.7 ± 9.31 kg (range 40–75 kg). The characteristics of all patients and the MICs of doripenem for the isolated pathogens are shown in Table 2. The population PK parameters of doripenem are shown in Table 1. All the tested covariates had no identifiable influence on the PK parameters, except between Vd and total clearance of doripenem (CL) (Appendix A). The PTAs for the two doripenem regimens achieving 40% T>MIC and 80% T>MIC at specific MICs are shown in Table 3 and Fig. 1. The values of Vd and CL of doripenem in these patients were 17.26 ± 1.83 L and 24.89 ± 1.63 L/h, respectively. For pathogens with a MIC of 1 ␮g/mL, the PTAs of achieving 40% T>MIC following administration of doripenem by a 1-h and 4-h infusion of 0.5 g q8h were 92.95% and 98.32%, respectively. For pathogens with a MIC of 2 ␮g/mL, the PTAs of achieving 40% T>MIC following administration of doripenem by a 1-h and 4-h infusion of 1 g q8h were 93.03% and 98.32%, respectively. For pathogens with a MIC of 4 ␮g/mL, the PTAs of achieving

Table 2 Characteristics of 11 patients with ventilator-associated pneumonia as well as the minimum inhibitory concentrations (MICs) of doripenem for the pathogens isolated from sputum. Patient

Age (years)

Sex

Weight (kg)

CLCr (mL/min)

Pathogen

MIC (␮g/mL)

APACHE II score

SOFA score

CPIS

1 2 3 4 5 6 7 8 9 10

73 39 25 47 80 40 47 41 41 64

M M M M M M F M M M

40 62 62 65 60 50 51 58 51 60.7

56 108 108 147 71 107 164 148 164 87

9 7 5 5 4 8 4 6 5 6

10 8 8 8 10 6 6 9 7 10

60

M

75

157

≥32 0.016 0.094 0.125 0.094 ≥32 0.016 0.19 – 0.016 0.023 0.023

23 28 15 17 15 21 21 17 17 25

11

Pseudomonas aeruginosa P. aeruginosa P. aeruginosa Acinetobacter baumannii P. aeruginosa Stenotrophomonas maltophilia P. aeruginosa A. baumannii NA Enterobacter cloacae Klebsiella pneumoniae K. pneumoniae

20

3

8

CLCr , creatinine clearance (determined by the Cockcroft–Gault method) [9]; APACHE, Acute Physiology and Chronic Health Evaluation; SOFA, Sepsis-related Organ Failure Assessment; CPIS, Clinical Pulmonary Infection Score; NA, not available.

Table 3 Probability of target attainment (PTA) for doripenem regimens achieving 40% T>MIC and 80% T>MIC in 11 patients with ventilator-associated pneumonia following administration by 1-h and 4-h infusion. MIC (␮g/mL)

PTA 40% T>MIC

PTA 80% T>MIC

1-h infusion

0.5 1 2 4 8

4-h infusion

1-h infusion

4-h infusion

0.5 g

1g

2g

0.5 g

1g

2g

0.5 g

1g

2g

0.5 g

1g

2g

99.91 92.95 48.79 5.47 0.08

99.99 99.83 93.03 49.13 5.61

100.00 100.00 99.88 93.15 48.72

100.00 98.32 70.44 13.33 0.29

100.00 99.99 98.32 70.54 13.35

100.00 100.00 99.99 98.34 70.7

56.82 17.66 1.46 0.00 0.00

85.72 56.41 17.83 1.22 0.01

96.02 86.03 56.57 17.87 1.41

91.14 50.97 7.30 0.15 0.00

99.71 90.92 51.14 6.91 0.10

100.00 99.71 91.21 50.86 6.81

MIC, minimum inhibitory concentration; T>MIC , time that concentrations in tissue and serum are above the MIC.

S. Jaruratanasirikul et al. / International Journal of Antimicrobial Agents 40 (2012) 434–439

Fig. 1. Probability of target attainment (PTA) for doripenem regimens achieving (A) 40% T>MIC and (B) 80% T>MIC at specific minimum inhibitory concentrations (MICs) in 11 patients with ventilator-associated pneumonia after administration of a 1-h infusion of 0.5 g every 8 h (q8h) (), a 4-h infusion of 0.5 g q8h (), a 1-h infusion of 1 g q8h (), a 4-h infusion of 1 g q8h (), a 1-h infusion of 2 g q8h (䊉) and a 4-h infusion of 2 g q8h (). The broken line represents 90% PTA. T>MIC , time that concentrations in tissue and serum are above the MIC.

40% T>MIC following administration of doripenem by a 1-h and 4h infusion of 2 g q8h were 93.15% and 98.34%, respectively. For pathogens with a MIC of 1 ␮g/mL, the PTA of achieving 80% T>MIC following administration of doripenem by a 1-h and 4-h infusion of 1 g q8h were 56.41% and 90.92%, respectively. For pathogens with a MIC of 2 ␮g/mL, the PTA of achieving 80% T>MIC following administration of doripenem by a 1-h and 4-h infusion of 2 g q8h were 56.57% and 91.21%, respectively. Both regimens were well tolerated and there were no reported adverse events. 4. Discussion The current study in critically ill patients with VAP found that the values of Vd and CL of doripenem were two-fold higher than the values obtained from a previous study [16]. A higher dosage than

437

the standard dosage regimen may be required for the treatment of severe infections in this patient population. ␤-Lactam antibiotics, especially carbapenems, are one of the most important and commonly prescribed drugs for the treatment of nosocomial infections in critically ill patients with VAP. These agents are hydrophilic antimicrobial agents that are characterised by their wide distribution into the extracellular fluid. Drug concentrations in the pulmonary epithelial lining fluid (ELF) for extracellular respiratory pathogens should be the primary determinants for therapeutic efficacy [17,18]. However, antibiotic concentrations in the ELF are very difficult to determine and the correlation between the PK/PD index in the ELF and antimicrobial effect is less well understood [19]. Therefore, serum concentrations are most commonly used as a surrogate measure for determining the PK/PD indices, and T>MIC is the best parameter that correlates with the bactericidal activity of ␤-lactams. High peak concentrations do not enhance the bactericidal activity of these agents and bacterial growth resumes rapidly when the level of antibiotics decreases to below the MIC [4,5]. Studies in animal infection models have shown that for most ␤-lactams, concentrations do not need to exceed the MIC for 100% of the dosing interval to achieve a significant antibacterial effect [4,5]. Bacteriostatic effects of carbapenems against Escherichia coli and P. aeruginosa in a murine thigh infection model are observed when serum drug concentrations are above the MIC for 20% of the dosing interval, whereas the T>MIC required for bactericidal activity is 40% of the dosing interval [20]. Moreover, optimum killing properties have been observed in critically ill patients when concentrations are maintained at 4× MIC, with higher concentrations providing little added benefit [21,22]. However, there is no consensus about which strategy between T>MIC and T>4×MIC is better. Several pathophysiological conditions, including variations in extracellular fluid and renal clearance in critically ill patients, are important factors determining the therapeutic success of hydrophilic antibiotics. The increase of Vd caused by the presence of extensive fluid extravasation as well as enhanced antimicrobial renal excretion result in decreased drug concentrations and may require a higher dosage of antimicrobial agents than standard dose recommendations [23]. Moreover, assessment of the pharmacokinetics of imipenem in critically ill patients with nosocomial pneumonia found that the Vd was higher than in historical controls, suggesting that the standard dosage of 0.5 g every 6 h of imipenem may result in lower concentrations in these patients [24]. The current study was conducted in very seriously ill patients with VAP due to Gram-negative bacilli in the ICU. The majority of patients had an APACHE II score ≥20 and a mean SOFA score ≥5, and more than 80% of them had a CPIS > 6. The values of Vd and CL of doripenem in these patients were two-fold higher than the values obtained from previous phase I studies with healthy volunteers as well as phase II and III studies in critically ill patients with complicated UTI or pyelonephritis or nosocomial pneumonia [16]. A possible explanation for these results is that the patients enrolled in this study had more severe illness than those from the previous study. A previous study has demonstrated that free drug is available for antimicrobial activity; however, doripenem has low protein binding (<10%) [3] and therefore the T>MIC both for free and total drug required for bactericidal effect is not much different. A previous PK/PD modelling of doripenem in normal volunteers and patients with renal dysfunction found that a 1-h infusion of 0.5 g of doripenem q8h achieved good coverage for Gramnegative bacilli with MICs of ≤1 ␮g/mL, and for the treatment of pathogens with higher MIC values a 4-h infusion of 1 g q8h achieved good coverage for pathogens with MICs of ≤8 ␮g/mL [25]. Another previous PK/PD target attainment analyses of doripenem in patients with UTIs and prostatitis also confirmed that prolonging the infusion time was a more effective strategy to achieve optimal

438

S. Jaruratanasirikul et al. / International Journal of Antimicrobial Agents 40 (2012) 434–439

pharmacodynamic exposure for pathogens with higher MIC than dose escalation [26]. However, the US Food and Drug Administration (FDA) was informing the public that a recent clinical trial with Doribax (doripenem) was stopped early because of significant safety concerns. This trial, which was evaluating the effects of Doribax on treatment of patients with VAP, demonstrated excess mortality and a numerically poorer clinical cure rate among subjects treated with Doribax compared with those treated with imipenem/cilastatin. In addition, for subjects infected with P. aeruginosa in the trial, the mortality rate for subjects in the doripenem treatment group was higher than in the imipenem/cilastatin treatment group [27]. A possible explanation of these results is that a fixed 7-day course of 4-h infusion of 1 g of doripenem q8h is inadequate to treat infection in critically ill patients compared with a fixed 10-day course of 1-h infusion of 1 g of imipenem/cilastatin q8h. In addition, a previous study of the intrapulmonary pharmacokinetics of doripenem in healthy adult subjects found that drug concentrations of this agent in ELF were much lower than the serum concentrations [28]. In the current study, a PK/PD study of doripenem in patients with VAP was examined, and a MCS was performed to determine the probability of attaining a specific PD target. The probability of a 4-h doripenem infusion regimen achieving a target of 40% T>MIC was superior to the probability of a 1-h infusion regimen. A high PTA (≥90%) of achieving 40% T>MIC for MICs of 1 ␮g/mL was observed when doripenem was administered by either a 1-h or 4-h infusion of 0.5 g q8h regimen. For pathogens with a MIC of 4 ␮g/mL, the high PTA (≥90%) was achieved when the dosage of doripenem was increased to a 2 g q8h regimen and, moreover, only a 4-h prolonged infusion of 2 g q8h regimen achieved >98% PTA of 40% T>MIC . A previous clinical study in febrile neutropenic patients was conducted to investigate the relationship between the PK/PD index of T>MIC of meropenem and clinical outcome. The results clearly demonstrated that an optimum clinical response was shown when the percentages of T>MIC of meropenem exceeded 75% of the dosing interval [29]. In the current study, the probability of a 4-h doripenem infusion regimen achieving a target of 80% T>MIC was much greater than the probability of a 1-h infusion regimen. For treatment of severe infections in neutropenic patients, the high PTA (≥90%) of achieving 80% T>MIC for MICs of 1 ␮g/mL and 2 ␮g/mL was observed when doripenem was administered by a 4-h infusion of 1 g q8h and 2 g q8h, respectively. From these data it appears that for the treatment of most severe infections in critically ill patients with altered pharmacokinetics, a more aggressive administration schedule of this agent may be required with a higher dosage than the standard dosage regimen. Although adverse events with the standard dosage regimen are uncommon, the care-giver should be aware of the risk of toxicity from a higher dosage regimen. In addition, the costs of increased dosage regimens of doripenem in critically ill patients and immunocompromised hosts are increased. During the doripenem administration in this study, no major adverse events related to this agent were observed. This study has some limitations that must be considered. First, the results of this study could be difficult to extrapolate to other situations because the low body weight of the patients could have had an effect on Vd and CL. Second, the small number of patients could be considered a potential limitation. However, in the absence of data from a larger sample size, a MCS based on a small number of patients such as in this study can be instructive in illuminating the effects of different dosing approaches [30]. Third, the Cockcroft–Gault method used for estimating CLCr is known to have certain limitations in severely ill patients [31]. Fourth, serum concentrations of doripenem in this study were measured at steady-state rather than during the early phase of therapy, which may be even more important in critically ill patients because, during the early phase of antibiotic therapy in severe sepsis the

increase of Vd from fluid resuscitation has an influence on antimicrobial concentrations. Fifth, four patients in the current study were neurosurgical patients who were admitted into the ICU. This could explain the high APACHE II scores of the patients, and the results of this study have some limitations to extrapolate to critically ill septic patients. In conclusion, the current population PK study found that the Vd and CL values of doripenem in critically ill patients with VAP were higher than the results obtained from a previous study. A high PTA (≥90%) of achieving 40% T>MIC for a MIC of 1 ␮g/mL was observed when doripenem was administered by either a 1-h or 4-h infusion of 0.5 g q8h. In addition, against pathogens less susceptible to doripenem with MICs > 1 ␮g/mL, the dosage of doripenem should be increased to 1 g q8h. Therefore, the dosage of doripenem for treatment of severe infections in this patient population should be higher than the standard dosage regimen. Acknowledgments Doripenem (DoribaxTM ) was generously donated by JanssenCilag Ltd. (Bangkok, Thailand). Doripenem standard powder was generously donated by Johnson & Johnson Pharmaceutical Research & Development (Raritan, NJ). The pharmaceutical companies were not involved with, and did not interfere with, data collection, interpretation of findings or preparation of the manuscript. The authors thank Mr David Patterson for checking the English of the manuscript. Funding: This work was supported by a faculty grant from the Faculty of Medicine, Prince of Songkla University (Hat Yai, Thailand). Competing interests: None declared. Ethical approval: This study was approved by the Ethics Committee of Songklanagarind Hospital (Prince of Songkla University, Hat Yai, Thailand). The study was conducted following the guidelines of the Declaration of Helsinki. Judgement reference no. EC. 53-281-14-1-1. Appendix A. Comparison of covariance matrix between the actual and simulated pharmacokinetic parametersa Parameter

Actual

Simulated

k12 –k21 k12 –ke k12 –Vd k12 –CL k21 –ke k21 –Vd k21 –CL ke –Vd ke –CL Vd –CL

0.626 −0.193 0.044 −0.063 −0.417 0.339 0.164 −0.585 −0.115 0.873

0.625 −0.192 −0.038 −0.140 −0.545 0.233 −0.011 −0.459 −0.015 0.895

k12 , intercompartmental transfer rate constant from compartment X1 to X2 ; k21 , intercompartmental transfer rate constant from compartment X2 to X1 ; ke , elimination rate constant from X1 ; Vd , volume of distribution; CL, total clearance. a The covariances are not statistically different at ˛ = 0.10.

References [1] Nicasio AM, Kuti JL, Nicolau DP. The current state of multidrug-resistant Gramnegative bacilli in North America. Pharmacotherapy 2008;28:235–49. [2] Talbot GH, Bradley J, Edwards Jr JE, Gilbert D, Scheld M, Bartlett JG. Bad bugs need drugs: an update on the development pipeline from the Antimicrobial Availability Task Force of the Infectious Diseases Society of America. Clin Infect Dis 2006;42:657–68. [3] Keam SJ. Doripenem: a review of its use in the treatment of bacterial infections. Drugs 2008;68:2021–57. [4] Craig WA. Interrelationship between pharmacokinetics and pharmacodynamics in determining dosage regimens for broad-spectrum cephalosporins. Diagn Microbiol Infect Dis 1995;22:89–96.

S. Jaruratanasirikul et al. / International Journal of Antimicrobial Agents 40 (2012) 434–439 [5] 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. [6] Hospital-acquired pneumonia in adults: diagnosis, assessment of severity, initial antimicrobial therapy, and preventive strategies. A consensus statement, American Thoracic Society, November 1995. Am J Respir Crit Care Med 1996;153:1711–25. [7] Roberts JA, Kwa A, Montakantikul P, Gomersall C, Kuti JL, Nicolau DP. Pharmacodynamic profiling of intravenous antibiotics against prevalent Gram-negative organisms across the globe: the PASSPORT Program—Asia-Pacific Region. Int J Antimicrob Agents 2011;37:225–9. [8] Taccone FS, Laterre PF, Dugernier T, Spapen H, Delattre I, Wittebole X, et al. Insufficient ␤-lactam concentrations in the early phase of severe sepsis and septic shock. Crit Care 2010;14:R126. [9] Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31–41. [10] File Jr TM. Recommendations for treatment of hospital-acquired and ventilatorassociated pneumonia: review of recent international guidelines. Clin Infect Dis 2010;51(Suppl. 1):S42–7. [11] Ikeda K, Ikawa K, Morikawa N, Kameda K, Urakawa N, Ohge H, et al. Quantification of doripenem in human plasma and peritoneal fluid by high-performance liquid chromatography with ultraviolet detection. J Chromatogr B Analyt Technol Biomed Life Sci 2008;867:20–5. [12] Mathews JH. Numerical methods for computer science, engineering and mathematics. London, UK: Prentice-Hall International, Inc.; 1987. [13] Li J, Rhinehart RR. Heuristic random optimization. Comput Chem Eng 1998;22:427–44. [14] Worakul N, Wongpoowarak W, Boonme P. Optimization in development of acetaminophen syrup formulation. Drug Dev Ind Pharm 2002;28:345–51. [15] Box GEP, Muller ME. A note on the generation of random normal deviates. Ann Math Statist 1958;29:610–1. [16] Nandy P, Samtani MN, Lin R. Population pharmacokinetics of doripenem based on data from phase 1 studies with healthy volunteers and phase 2 and 3 studies with critically ill patients. Antimicrob Agents Chemother 2010;54: 2354–9. [17] Kikuchi E, Kikuchi J, Nasuhara Y, Oizumi S, Ishizaka A, Nishimura M. Comparison of the pharmacodynamics of biapenem in bronchial epithelial lining fluid in healthy volunteers given half-hour and three-hour intravenous infusions. Antimicrob Agent Chemother 2009;53:2799–803. [18] Baldwin DR, Honeybourne D, Wise R. Pulmonary disposition of antimicrobial agents: methodological considerations. Antimicrob Agents Chemother 1992;36:1171–5. [19] Mouton JW, Ambrose PG, Kahlmeter G, Wikler M, Craig WA. Applying pharmacodynamics for susceptibility breakpoint selection and susceptibility testing.

[20] [21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

439

In: Nightingale CH, Ambrose PG, Drusano GL, Murakawa T, editors. Antimicrobial pharmacodynamics in theory and clinical practice. New York, NY: Informa Healthcare; 2007. p. 21–44. Drusano GL. Prevention of resistance: a goal for dose selection for antimicrobial agents. Clin Infect Dis 2003;36(Suppl. 1):S42–50. 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. Craig WA. Basic pharmacodynamics of antibacterials with clinical applications to the use of ␤-lactams, glycopeptides, and linezolid. Infect Dis Clin North Am 2003;17:479–501. Udy AA, Varghese JM, Altukroni M, Briscoe S, McWhinney B, Ungerer J, et al. Subtherapeutic initial ␤-lactam concentrations in select critically ill patients: association between augmented renal clearance and low trough drug concentrations. Chest 2011, http://dx.doi.org/10.1378/chest.11-1671 Epub ahead of print. McKindley DS, Boucher BA, Hess MM, Croce MA, Fabian TC. Pharmacokinetics of aztreonam and imipenem in critically ill patients with pneumonia. Pharmacotherapy 1996;16:924–31. Van Wart SA, Andes DR, Ambrose PG, Bhavnani SM. Pharmacokinetic–pharmacodynamic modeling to support doripenem dose regimen optimization for critically ill patients. Diagn Microbiol Infect Dis 2009;63:409–14. Ikawa K, Morikawa N, Uehara S, Monden K, Yamada Y, Honda N, et al. Pharmacokinetic–pharmacodynamic target attainment analysis of doripenem in infected patients. Int J Antimicrob Agents 2009;33:276–9. US Food and Drug Administration. FDA statement on recently terminated clinical trial with Doribax (doripenem). http://www.fda.gov/Drugs/ DrugSafety/ucm285883.htm [accessed 12.04.12]. Justo J, Gotfried MH, Deyo K, Fischer P, Danziger LH, Rodvold KA. Doripenem intrapulmonary pharmacokinetics in healthy adult subjects. In: 51st Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), 17–20 September 2011. Chicago: ASM Press; 2011 [poster A1-1748]. Ariano RE, Nyhlen A, Donnelly JP, Sitar DS, Harding GKM, Zelenitsky SA. Pharmacokinetics and pharmacodynamics of meropenem in febrile neutropenic patients with bacteremia. Ann Pharmacother 2005;39: 32–8. Roberts JA, Kirkpatrick CMJ, Lipman J. Monte Carlo simulations: maximizing antibiotic pharmacokinetic data to optimize clinical practice for critically ill patients. J Antimicrob Chemother 2011;66:227–31. Martin JH, Fay MF, Udy A, Roberts J, Kirkpatrick C, Ungerer J, et al. Pitfalls of using estimations of glomerular filtration rate in an intensive care population. Intern Med J 2010, http://dx.doi.org/10.1111/j.1445-5994.2010.02160.x Epub ahead of print.