Population Pharmacokinetic Modeling of Piperacillin/Tazobactam in Healthy Adults and Exploration of Optimal Dosing Strategies
1
Department of Physical Medicine and Rehabilitation, Myongji Hospital, Goyang 10475, Republic of Korea
2
Division of Clinical Pharmacology, Chonnam National University Hospital, Gwangju 61469, Republic of Korea
3
Division of Hematology-Oncology, Department of Internal Medicine, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Anyang 14066, Republic of Korea
4
Department of Clinical Pharmacology, Hallum University Sacred Heart Hospital, Hallym University College of Medicine, Anyang 14066, Republic of Korea
*
Author to whom correspondence should be addressed.
Pharmaceuticals 2025, 18(8), 1124; https://doi.org/10.3390/ph18081124 (registering DOI)
Submission received: 16 June 2025
/
Revised: 14 July 2025
/
Accepted: 24 July 2025
/
Published: 27 July 2025
(This article belongs to the Special Issue Therapeutic Drug Monitoring and Adverse Drug Reactions: 2nd Edition)
Abstract
Background/Objectives: Current dosing recommendations for piperacillin/tazobactam suggest adjustments only for patients with creatinine clearance (CrCl) below 40 mL/min, potentially neglecting the variability in drug exposure among patients with a CrCl greater than 40 mL/min. This study aimed to develop a population pharmacokinetic (PK) model for piperacillin/tazobactam and explore optimal dosage regimens tailored by renal function and pathogen susceptibility. Methods: Twelve healthy adults received a single intravenous dose of piperacillin/tazobactam (4 g/0.5 g). Population PK models were developed using nonlinear mixed-effects modeling. Monte Carlo simulations were conducted to identify optimal dosing regimens across various renal functions and MIC levels, guided by pharmacodynamic targets defined as the percentage of time that free drug concentrations exceed the minimum inhibitory concentration (fT>MIC). Results: PK profiles of both drugs were best described by two-compartment models. Estimated glomerular filtration rate (eGFR) adjusted by body surface area and body weight were identified as significant covariates influencing drug clearance and peripheral volume of distribution. Simulations showed that the standard dosing regimen (4/0.5 g q6h with 30 min infusion) achieved a 90% probability of target attainment (PTA) for 50%fT>MIC at MIC values up to 4 mg/L in patients with normal renal function. However, this regimen often did not achieve a 90% PTA for stringent targets (100%fT>MIC, 100%fT>4MIC) or higher MICs, particularly in patients with eGFR ≥ 130 mL/min. Conclusions: These findings suggest current dosing regimens may be inadequate and highlight the potential of alternative strategies, such as extended or continuous infusion, which warrant further investigation in clinical populations to optimize therapeutic outcomes.
1. Introduction
Piperacillin/tazobactam is a penicillin-class β-lactam/β-lactamase inhibitor combination that has received approval from the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) for the treatment of a wide range of bacterial infections, including intra-abdominal infections, skin and skin structure infections, urinary tract infections, female pelvic infections, nosocomial pneumonia, and community-acquired pneumonia [1,2]. It exhibits broad-spectrum antibacterial activity against aerobic and anaerobic Gram-positive and Gram-negative bacteria. It is frequently used for empirical or definitive treatment of critically ill patients with sepsis [3,4,5].
Early clinical data have affirmed the efficacy of this fixed-dose combination (at a dose ratio of 8:1) for managing moderate to severe polymicrobial infections, consistent with its in vitro activity [6]. Johnson and colleagues examined its pharmacokinetics (PK) in patients with renal disease and found that the half-life and area under the plasma drug concentration-time curve of piperacillin and tazobactam concomitantly increase as creatinine clearance (CrCl) decreases. Based on these findings, they recommended lower daily doses for patients with a CrCl below 40 mL/min [7]. In clinical practice, such as in the management of adults with nosocomial pneumonia, piperacillin/tazobactam is typically prescribed at a fixed dose of 4/0.5 g q6h for patients with a CrCl greater than 40 mL/min [1,2]. However, this fixed regimen does not account for the variability of renal function beyond this threshold, which may result in significant differences in drug exposure. These differences may affect pharmacodynamic (PD) outcomes and efficacy, increasing the risk of resistance. Critically ill patients, such as those with sepsis, often experience pathophysiological changes like fluid shifts from leaky capillaries and altered protein binding, which complicate drug disposition and dosing [8]. These shifts affect the volume of distribution (Vd) and drug clearance (CL). Decreased renal function can reduce CL, while an increase in cardiac output may enhance it. Given these complexities, model-informed precision dosing should be employed to account for individual pathophysiological conditions, including decreased, normal, or increased renal function.
Previous studies have shown that the therapeutic efficacy of piperacillin/tazobactam administered at a fixed dose of 4/0.5 g q6h can vary based on renal function, particularly in patients with a CrCl exceeding 40 mL/min. Efficacy was evaluated by the probability of target attainment (PTA), defined as the total percentage of a 24 h period during which the concentration of the free (unbound to protein) drug surpasses the minimum inhibitory concentration (MIC) under steady-state conditions (fT>MIC) [9,10,11]. Asín-Prieto et al. established a population PK model for critically ill patients undergoing continuous renal replacement therapy, in which piperacillin/tazobactam was administered q6h over 20 min [12]. Their simulations were aimed at achieving a 90% PTA of 100%fT>MIC and indicated that with a CrCl of 50 mL/min, the breakpoints were 8 mg/L, which decreased to 2 mg/L as the CrCl increased to 100 mL/min. Sime et al. developed a PK model for patients with febrile neutropenia [13]. Using this model to simulate a fixed-dose administration over 30 min, they aimed for a 90% PTA of 50%fT>MIC and found that at a CrCl of 40 mL/min, the breakpoints were at 16 mg/L, which decreased to 4 mg/L at 80 mL/min and further to 1 mg/L at 120 mL/min. Ishihara et al. developed a PK model for elderly patients with pneumonia and conducted simulations targeting a 90% PTA of 50%fT>MIC with a 1 h infusion q6h [14]. The results indicated that at a CrCl of 50 mL/min, the breakpoints were at 16 mg/L, but were reduced to 8 mg/L when CrCl increased to 60 mL/min.
While the preceding studies highlight the need for dosing optimization, they were conducted in clinically complex patient populations where it is difficult to isolate the effect of renal function from other disease-related confounding factors. This necessitates a foundational study in healthy adults to establish a PK baseline free from disease-related confounding factors. Moreover, as existing PK data from healthy volunteers originate primarily from Western populations, and ethnic differences in drug disposition are well-recognized, a population-specific model for healthy Korean adults was required to serve as a more appropriate reference. This study was therefore designed to address this dual knowledge gap.
The objectives of this study were to develop a population PK model for piperacillin/tazobactam in healthy adults and to perform Monte Carlo simulations using this model to suggest optimal dosing and administration strategies for adult patients with a CrCl of 40 mL/min or greater.
2. Results
2.1. Participants
The demographic and clinical characteristics of the 12 healthy adult subjects are listed in Table 1. All baseline laboratory values were within normal clinical ranges, confirming their healthy status. The median body weight of our Korean subjects was 61.7 kg, and key renal function metrics such as CrCl and estimated glomerular filtration rates (eGFR) confirmed normal renal function across the cohort. During the stepwise covariate analysis, various forms of the CKD-EPI equation utilizing creatinine, cystatin C, or both were evaluated as potential covariates for clearance [15]. As these estimators yielded broadly consistent values in our population, only representative variables with potential clinical relevance were included in Table 1. All participants completed the study without any adverse drug reactions.
2.2. Population Pharmacokinetic Analysis
A total of 84 plasma samples (Figure 1) were used to establish a population PK model for piperacillin/tazobactam. The concentration-time profiles for both piperacillin and tazobactam were best described by a two-compartment model. The structural PK parameters for the two-compartment model were CL, Vd for the central compartment (V1), Vd for the peripheral compartment (V2), and intercompartmental CL between V1 and V2 (Q), as shown in Table 2 and Table 3. The BSA-adjusted eGFR, calculated using the CKD-EPI equations and creatinine levels, was identified as a significant covariate for CL in the final PK models for both piperacillin and tazobactam (Table 2 and Table 3). In addition, body weight was a significant covariate for the V2 in these models. For piperacillin, lean body mass (LBM) was also a significant covariate for Q in the final PK model. The incorporation of eGFR as a covariate on CL significantly improved the model fit for piperacillin; its removal from the final model increased the OFV by 16.414 and reduced the associated interindividual variability (IIV) from 13.2% to 7.17%. Similarly, the inclusion of CrCl on CL was significant for tazobactam, increasing OFV by 13.318 upon removal and reducing IIV from 17.6% to 6.95%. Details of the covariate selection process are provided in Appendix A (Table A1 and Table A2). During covariate evaluation, WT was significantly associated with V2 for both piperacillin and tazobactam, with notable reductions in OFV. In contrast, the inclusion of WT as a covariate on V1 did not yield statistically significant improvements. In the final PK model for tazobactam, IIV on the V1 was initially tested; however, due to its high relative standard error (RSE > 70%) and negligible contribution to model fit improvement, this parameter was excluded from the final model. Conversely, for piperacillin, the IIV on V1 showed acceptable precision (RSE: 28.7%, shrinkage: 19.1%) and was retained in the final model (Table 2 and Table 3).
Figure 2 presents diagnostic goodness-of-fit plots for the final PK model of piperacillin and tazobactam. Plots of conditional weighted residuals (CWRES) versus time and versus population predicted concentrations showed largely uniform around the x-axis or the line of identity (y = x), suggesting that the structural models were correctly adjusted and without bias. Minor deviations for two data points around 4–5 h in panels (a) and (e) were observed, but the overall distribution of CWRES near zero indicated an acceptable model fit. The time courses for the individual fit plots are shown in Figure A1 and Figure A2 for piperacillin and tazobactam, respectively. The visual predictive checks (VPC) for piperacillin and tazobactam are shown in Figure A3. The measured 10th, 50th, and 90th percentiles fell within the 95% confidence intervals (CI) of the respective simulated percentiles, indicating that the final PK models adequately described the observed concentration variability and had good predictive performance.
2.3. Dosage Simulation
Simulation results for empirical dosing in subjects with normal renal function showed variable PTA across the different infusion durations and targeted pathogens (Figure 3). For clinical context, according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) database [16], the clinical breakpoint for susceptibility is an MIC ≤ 8 mg/L for Enterobacterales (including Escherichia coli and Klebsiella pneumoniae) and an MIC ≤ 16 mg/L for Pseudomonas aeruginosa.
The standard dosing regimen of piperacillin/tazobactam (4 g/0.5 g q6h with a 0.5 h infusion) achieved adequate PTA (>90%) at lower MIC values. However, a pronounced decline in PTA was observed at elevated MIC values, particularly for E. coli and K. pneumoniae, even with prolonged infusion durations. For P. aeruginosa, extending the infusion duration to ≥3 h significantly improved PTA, maintaining PTA > 90% up to an MIC of 16 mg/L. In contrast, for all three pathogens, shorter infusion times (≤1 h) consistently failed to achieve the 90% PTA threshold at MIC values of 8 mg/L or higher. These results underscore the limitations of current label-based empirical dosing regimens against pathogens with higher MIC values and highlight the potential advantage of extended infusion strategies. As shown in Figure A4, the cumulative fraction of response (CFR) increased with longer infusion times across all pathogens evaluated. The improvement was most pronounced in K. pneumoniae, where CFR rose from approximately 15% at 0.5 h to over 75% at 4 h. In contrast, E. coli and P. aeruginosa exhibited higher baseline CFRs (>65% at 0.5 h), and their response curves showed a more gradual ascent, reaching near-maximal levels (>90%) at 4 h.
The second simulation study provided PTA results for targeted dosing regimens across various renal function groups, infusion strategies, and MIC values (Figure 4, Figure A5 and Figure 5). For less stringent PD targets (50%fT>MIC and 50%fT>4MIC), shorter infusions (0.5 h) were generally sufficient to achieve adequate PTA at lower MIC values. However, their effectiveness diminished substantially at higher MICs, a challenge particularly pronounced in subjects with eGFR ≥ 130 mL/min, where even high doses often failed to achieve 90% PTA for MICs ≥ 4 mg/L (Figure 4 and Figure A5). Extending the infusion duration from 0.5 h to 3 h consistently enhanced PTA across all renal function categories, especially at moderate MICs.
Achieving the stricter PD targets (100%fT>MIC and 100%fT>4MIC) was considerably more challenging. For patients with normal renal function (e.g., eGFR 90–130 mL/min), high-dose intermittent regimens, such as 6 g q6h infused over 3 h, were required to reliably cover MICs up to only 2 mg/L for the 100%fT>MIC target (Figure 4). In contrast, continuous infusion strategies demonstrated superior performance; a daily dose of 16 g via continuous infusion achieved nearly 100% PTA for the same target up to an MIC of 16 mg/L (Figure 5). Table A3 and Table A4 summarize the recommended dosing regimens derived from these simulations. Collectively, these results reinforce the need for personalized, model-informed dosing strategies based on renal function and pathogen susceptibility to optimize therapeutic outcomes.
To aid in understanding our results, a user-friendly Shiny app has been developed (available at https://dhlee.shinyapps.io/pthv/).
3. Discussion
This study addresses a notable gap in the literature by conducting a population PK analysis of piperacillin/tazobactam in healthy Korean adults, providing foundational insights for future clinical applications. Our population PK analysis demonstrated that the PK of both piperacillin and tazobactam were best described by two-compartment models. The final models incorporated several significant covariates: for both drugs, renal function (eGFR, calculated by the CKD-EPI formula) influenced CL and WT affected V2, and LBM was specifically identified as a key determinant of Q for piperacillin.
Although several renal function estimators were evaluated, including creatinine-based, cystatin C-based, and combined equations, the final selection of the creatinine-based CKD-EPI equation was driven by statistical performance in the model. Given the limited sample size, it is plausible that different renal function markers might have emerged as significant in a larger or more heterogeneous cohort. The statistically supported assignment of WT specifically to V2 rather than V1 might appear somewhat counterintuitive. Typically, V1 is more directly associated with body size metrics, reflecting the initial distribution space of the drug. However, given the hydrophilic properties of piperacillin and tazobactam, the peripheral compartment likely reflects distribution into extravascular spaces such as interstitial fluids, muscle, and skin tissues, volumes strongly correlated with body weight. Thus, this data-driven covariate assignment appears physiologically plausible. The stepwise covariate modeling provided strong support for including LBM as a covariate on Q for piperacillin (ΔOFV = 14.798), a finding that, while not typical, was statistically robust. Initially, theory-based allometric scaling [17] involving total body weight or LBM across all PK parameters was attempted, but these broader approaches did not improve model performance. A mechanistic rationale is that LBM closely reflects lean tissues and extracellular fluid compartments, significantly relevant to hydrophilic antibiotics like piperacillin. Nonetheless, these findings, particularly the novel association of LBM with Q, must be interpreted with caution. Although the parameter was statistically robust and passed bootstrap validation, the limited sample size creates an inherent risk of overfitting. Therefore, this specific finding requires external validation in larger, more diverse populations before its clinical significance can be established.
In this study, typical values for piperacillin CL and steady-state volume of distribution (Vss = V1 + V2) were 11.2 L/h and 8.83 L, respectively, corresponding to 0.174 L/h/kg and 0.148 L/kg when normalized by body weight. These findings are slightly lower for clearance and notably lower for Vss compared with previously reported noncompartmental analyses in healthy adults. For example, Bryson and Brogden summarized piperacillin’s CL and Vss as approximately 14.5 L/h (0.181 L/h/kg) and 15.0 L (0.188 L/kg), respectively, assuming an average weight of 80 kg [6]. Similarly, Daniel and Krop reported CL and Vss values of 15.5 L/h (0.194 L/h/kg) and 16.7 L (0.209 L/kg), respectively, under the same assumption [18].
Population PK studies have reported a range of CL and Vss values. Bulitta et al. compared the PK of piperacillin in 8 Caucasian patients with cystic fibrosis to that in 26 healthy volunteers. In their study, the CL for the healthy individuals was 10.4 L/h (0.146 L/h/kg), and the Vss was 11.8 L (0.166 L/kg) [19]. Bulitta et al. conducted another study on 4 healthy Caucasian adults. In their study, the CL was 10.1 L/h (0.144 L/h/kg), and the Vss was 12.7 L (0.181 L/kg) [20]. Felton et al. developed a PK model based on data from 53 critically ill patients and 25 healthy volunteers. In their study, the CL for healthy individuals was 11.45 L/h (0.142 L/h/kg), and the Vss was 8.18 L (0.102 L/kg) [21]. It is also important to consider the demographic differences between these cohorts, particularly body weight. The median body weight of our Korean subjects was 61.7 kg, which is considerably lower than the mean or median body weights of the healthy Caucasian volunteers in the studies by Bulitta et al. (71.1 kg and 77.5 kg, respectively) and Felton et al. (81 kg) [19,20,21]. Notably, despite nearly 20 kg lower median body weight compared to the subjects in the Felton et al. study, the Vss was remarkably similar (8.99 L vs. 8.18 L). This suggests that factors beyond body weight, such as body composition or ethnicity, may significantly influence piperacillin’s Vd.
In patients with normal or above-normal renal function, the Vss was generally higher than in our study. This variation is further illustrated by population PK studies on critically ill sepsis patients. Roberts et al. found a median CL and Vss for piperacillin at 17.1 L/h (0.214 L/h/kg) and 25 L (0.313 L/kg), respectively [22]. Similarly, Udy et al. reported a median CrCl of 122 mL/min and a median weight of 88.4 kg, with CL and Vss of 16.3 L/h (0.184 L/h/kg) and 38.7 L (0.438 L/kg), respectively [23]. Sime et al. reported on a population PK study involving critically ill sepsis patients with a median CrCl of 94 mL/min/1.73 m2 and a median weight of 78 kg, where CL and Vss for piperacillin were 18.0 L/h (0.231 L/h/kg) and 18.3 L (0.235 L/kg), respectively [13].
Differences in Vss are also observed in studies involving Asian populations, suggesting factors beyond ethnicity, such as clinical status, play a significant role. Direct comparisons are challenging, but various population PK studies illustrate this point. For example, Kim et al. conducted a study on critically ill Korean patients with a median CrCl of 53.1 mL/min and a median weight of 54 kg. They reported a clearance of 5.05 L/h (0.0904 L/h/kg) and a Vss of 22.8 L (0.422 L/kg) [24]. Ishihara et al. studied elderly Japanese patients with an average CrCl of 38.0 mL/min and an average weight of 45.5 kg. They reported CL and Vss values of 4.58 L/h (0.101 L/h/kg) and 12.4 L (0.271 L/kg), respectively [14]. Chen et al. evaluated patients with nosocomial infections who had an average CrCl of 68.7 mL/min and an average weight of 61.1 kg. The CL and Vss were 9.14 L/h (0.150 L/h/kg) and 12.2 L (0.200 L/kg), respectively [25].
Synthesizing these comparisons reveals the central contribution of our work. The Vss in our healthy Korean subjects was markedly lower than that reported in patient populations, which was often two- to four-fold greater. This stark difference, likely driven by disease-state pathophysiology, underscores the necessity of establishing an “uncontaminated” baseline from healthy individuals to accurately quantify the effects of illness on drug disposition. Moreover, while other healthy-volunteer studies exist, they predominantly feature Caucasian populations. As ethnic differences can influence PK, our study addresses a critical gap by establishing the first population PK model specifically in healthy Korean adults. This work therefore provides a dual contribution: a clean baseline free from disease-related confounders and a population-specific foundation essential for developing tailored therapeutic strategies for Korean and East Asian patients.
The adequacy of current piperacillin/tazobactam dosing recommendations for patients with CrCl ≥ 40 mL/min was further examined through simulations, along with an evaluation of alternative dosing regimens stratified by narrower renal function intervals. In addition to the most used PK/PD target (50%fT>MIC), simulations examined more stringent targets suitable for severe infections, including 50%fT>4MIC, 100%fT>MIC, and 100%fT>4MIC. The target of 50%fT>MIC originated from efficacy observations in mouse models infected with Streptococcus pneumoniae and patients with otitis media caused by S. pneumoniae and Haemophilus influenzae [9]. This target is recognized as a critical determinant of clinical success [26,27,28]. Additional research in ICU populations demonstrated that higher fT>MIC correlates with improved bacteriological and clinical outcomes [29], promoting the adoption of these more rigorous thresholds in multiple studies [24,30,31,32].
The observed superiority of extended and continuous infusion strategies in our simulations is directly explained by the time-dependent pharmacodynamic nature of piperacillin. As a β-lactam antibiotic, its bactericidal activity is not primarily dependent on achieving high peak concentrations but rather on the cumulative percentage of time the free drug concentration remains above the MIC during the dosing interval [9,10]. Standard short infusions generate high peaks, but drug concentrations decline rapidly, thus shortening the duration of effective exposure. In contrast, extended infusion lowers the peak concentration but prolongs the time above the MIC, thereby maximizing the %fT>MIC for a given daily dose. Continuous infusion represents the ultimate application of this principle by maintaining a constant drug concentration, which explains its superior ability to achieve stringent targets (e.g., 100% fT>MIC), as demonstrated in Figure 5.
The first simulation assessed FDA and EMA dosing recommendations for piperacillin/tazobactam using population PK modeling and Monte Carlo simulations. Results showed that the standard regimen (4 g/0.5 g q6h, 0.5 h infusion) achieved adequate PTA (≥90%) at lower MICs for common nosocomial pneumonia pathogens (Figure 3). However, at elevated MIC levels, especially with E. coli and K. pneumoniae, significant reductions in PTA were observed even when infusion durations were prolonged. Conversely, extended infusions (≥3 h) notably improved PTA for P. aeruginosa, maintaining adequate attainment at MICs up to 16 mg/L. These findings highlight potential limitations of the current standard dosing approach and underscore the value of extended infusion durations, particularly against pathogens with higher MIC values. This trend was further supported by the CFR analysis (Figure A4), which demonstrated that prolonged infusion times enhanced the cumulative probability of achieving pharmacodynamic targets across all pathogens. Notably, K. pneumoniae showed the greatest increase in CFR—from approximately 15% at 0.5 h to over 75% at 4 h—largely due to the more stringent PD target applied (100%fT>MIC). In contrast, E. coli and P. aeruginosa, for which lower targets of 77% and 50% fT>MIC were used respectively, exhibited higher CFRs even at shorter infusion durations. These differences emphasize that CFR outcomes are influenced not only by infusion strategy but also by the choice of PD target, which in this study varied across pathogens by design.
In the second dosing simulations, both intermittent and continuous infusion regimens were evaluated across a wide range of renal function categories (Figure 4, Figure 5 and Figure A5). While shorter infusions (0.5 h) were generally sufficient for less stringent PD targets (50%fT>MIC and 50%fT>4MIC) at lower MIC values, their effectiveness diminished substantially at higher MICs, especially in the presence of augmented renal clearance (ARC) [33], a condition characterized by an eGFR of 130 mL/min or higher (Figure 4 and Figure A5). Extending infusion durations to 3 h significantly enhanced PTA across all renal function categories, especially at intermediate MIC values. Achieving more stringent PD targets (100%fT>MIC and 100%fT>4MIC) required higher dosing and longer infusion durations, especially for patients with eGFR ≥ 40 mL/min.
Continuous infusion regimens consistently provided higher PTA compared to intermittent infusions across all renal function groups and MIC values (Figure 5). These findings emphasize the clinical importance of individualized dosing strategies based on renal function and pathogen susceptibility, supporting the adoption of model-informed precision dosing approaches to optimize therapeutic efficacy and patient outcomes. These simulation data provide quantitative support for refining current clinical dosing practices of piperacillin/tazobactam, potentially improving therapeutic outcomes in critically ill patients with nosocomial infections.
Our study’s PK/PD breakpoints (the highest MIC at which the target PTA was achieved) showed significant differences compared to other studies that applied similar dosing regimens to patients with comparable renal functions. For a PK/PD index of 50%fT>MIC, we observed that patients with normal renal function could achieve a PTA of ≥90% up to an MIC of 4 mg/L with a 4 g q6h 0.5 h infusion, and up to an MIC of 32 mg/L with a 4 g q6h 3 h or 4 h extended infusion (Figure 3). Conversely, Patel et al. noted that for patients with a eGFR of 100 mL/min, the treatment goal was met up to an MIC of 1 mg/L with a 0.5 h infusion, and up to an MIC of 8 mg/L with a 4 h extended infusion [26]. In our assessment for a PK/PD index of 100%fT>MIC in patients with an eGFR of 90–130 mL/min, a 0.5 h infusion of 4 g q6h did not meet the treatment goal at an MIC of 0.5 mg/L. However, extending the infusion to 3 h allowed achievement of the goal up to an MIC of 2 mg/L (Figure 4), while a 16 g/day continuous infusion met the goal up to an MIC of 32 mg/L (Figure 5). Asín-Prieto et al. found that in patients with a CrCl of 100 mL/min, the treatment goals were achieved up to a MIC of 2 mg/L with a 4 g q6h 20 min infusion, up to 8 mg/L with a 4 h extended infusion, and up to 16 mg/L with a 16 g/day continuous infusion [12]. Klastrup et al. also reported that in patients with a CrCl of 80–130 mL/min, the treatment goal was achieved up to a MIC of 16 mg/L with a 16 g/day continuous infusion [34]. In our study, the relatively low Vd observed in our healthy subjects likely contributes to the challenge of maintaining concentrations above MIC for 100% of the interval with intermittent infusions, resulting in lower PK/PD breakpoints for this stringent target due to a shorter half-life associated with the relatively lower Vd observed in healthy subjects compared to patients with a larger Vd. These findings underscore key limitations of the current standard dosing regimen, underscoring the potential advantages of extended infusion durations, especially for pathogens with elevated MIC values.
This study had several limitations. First, the relatively small sample size of 12 healthy adults, despite employing a rich sampling strategy, limited our ability to precisely estimate IIV and directly generalize findings to diverse patient populations, particularly the critically ill. Second, the narrow range of baseline renal function within our healthy adult participants significantly restricts the clinical applicability of the PK models and dosing recommendations, underscoring the need for validation in broader populations with impaired or augmented renal clearance. Third, the unbound fraction of piperacillin was assumed to be a fixed value of 0.7, which does not account for the considerable variability observed in patient populations, especially among the critically ill [35,36]. While such variability could theoretically influence target attainment, previous sensitivity analyses have demonstrated that the PTA for piperacillin remains robust to moderate fluctuations in protein binding [37]. This supports the validity of our approach despite this simplifying assumption. Finally, during the second simulation, lean body mass and body weight were held constant at median values. While this simplified interpretation by clearly isolating renal function effects, it may limit the generalizability to a more diverse patient population. Therefore, our simulation results represent a foundational PK baseline rather than directly applicable clinical dosing guidelines, highlighting the necessity for dedicated PK studies in patient populations exhibiting relevant pathophysiological changes.
In conclusion, this study demonstrated that current label-recommended dosing regimens for piperacillin/tazobactam may fail to provide sufficient therapeutic exposure in patients with eGFR ≥ 40 mL/min, particularly those with ARC. A population PK model incorporating eGFR, LBM, and body weight effectively predicted piperacillin/tazobactam PK in healthy adults. Monte Carlo simulations demonstrated that individualized dosing strategies tailored by detailed renal function stratification significantly enhance the probability of achieving target PK/PD indices. Therefore, to enhance clinical efficacy and minimize the risk of resistance, clinicians should consider integrating renal function-based, model-informed precision dosing strategies, including prolonged or continuous infusion for patients with eGFR ≥ 40 mL/min, into clinical practice. Although the simulation results suggest that higher dosing regimens could enhance the probability of achieving PD targets, such intensified regimens inherently carry risks of toxicity and adverse effects. Patients with borderline renal function or significant comorbidities might be particularly susceptible. Therefore, any clinical application of these findings must carefully balance efficacy with safety, incorporating therapeutic drug monitoring and rigorous clinical surveillance. Ultimately, validation through targeted PK studies in actual patient populations, especially critically ill patients, remains essential to refine these preliminary dosing insights and ensure their safe and effective clinical translation.
To build upon these foundational findings and address the limitations of this study, future research should proceed in several key directions. First, large-scale population PK studies in diverse patient populations—including critically ill Korean and East Asian patients with conditions such as sepsis or nosocomial pneumonia—are needed to quantify the impact of disease-related pathophysiology on drug disposition. Second, these studies should include participants across the full spectrum of renal function, from severe impairment to augmented clearance, to validate and refine the proposed dosing regimens. Third, direct measurement of unbound piperacillin concentrations should be incorporated to account for variability in protein binding, particularly in patients with hypoalbuminemia. Ultimately, prospective, randomized controlled trials are essential to evaluate the safety and efficacy of extended or continuous infusions compared to standard dosing and to confirm their clinical and microbiological benefits.
4. Materials and Methods
4.1. Participants
The study received approval from the Institutional Review Board at Hallym University Sacred Heart Hospital (IRB No. 2022-08-006, approved on 25 October 2022) and was conducted in compliance with the Declaration of Helsinki. This study was registered with the Clinical Research Information Service, operated by the National Institute of Health of the Korea Disease Control and Prevention Agency. The registration number is KCT0009855. More information can be found on the website: https://cris.nih.go.kr/cris/. The study was conducted in January 2023 at the Clinical Trial Center of the same institution. The inclusion criteria included: (1) Individuals aged between 19 and 55 at the time of screening; (2) Individuals free from congenital or chronic health conditions as confirmed by a medical evaluation; and (3) Individuals deemed eligible after comprehensive health screenings that encompassed medical history reviews, vital signs checks, physical examinations, hematological and biochemical blood tests, urinalysis, and other diagnostic tests. The exclusion criteria were: (1) Individuals with significant medical conditions affecting various systems, such as gastrointestinal, cardiovascular, respiratory, endocrine, hepatobiliary, hematologic-oncologic, musculoskeletal, renal, neurological, psychiatric, immunological, urological, ophthalmological, otolaryngological, or genetic disorders; (2) Individuals with past health issues that might influence the PK of the drugs, including liver or kidney diseases; (3) Individuals allergic to piperacillin/tazobactam or who have experienced adverse reactions to it; (4) Individuals testing positive for hepatitis B surface antigen, hepatitis C virus antibodies, HIV antigen or antibodies, or syphilis; and (5) Women who are pregnant, breastfeeding, or who may become pregnant.
4.2. Study Design
Eligible participants received an intravenous infusion of 4/0.5 mg piperacillin/tazobactam, prepared in 100 mL of saline solution and administered over 30 min. Venous blood samples (6 mL) were collected into tubes containing EDTA at predetermined time points to assess the PK characteristics. The planned sampling times were immediately before dosing and 0.5, 0.75, 1, 2, 3, and 6 h after beginning the infusion. Plasma concentrations of piperacillin/tazobactam were measured using a validated liquid chromatography-tandem mass spectrometry assay [24].
4.3. Population PK Analysis
The PK parameters for piperacillin/tazobactam were evaluated using nonlinear mixed-effects models with the assistance of NONMEM software (version 7.5, ICON Clinical Research LLC, North Wales, PA, USA). The first-order conditional estimation with interaction method was used for parameter estimation and incorporated both fixed and random effects. For the PK modeling of piperacillin/tazobactam, one-, two-, and three-compartment models were evaluated. Each model operated under first-order kinetics, except in the case of zero-order infusion processes. The models used the formula θi = θ × exp(ηi) to define each parameter, where θ is the typical value, θi is the individual parameter value, and ηi represents the IIV, which is assumed to follow a normal distribution with zero mean and variance ω2. Residual variability, characterized by a normal distribution with zero mean and variance σ2, was assessed using additive, proportional, and combined additive-proportional error models. During population PK model development, covariance structures between key PK parameters (e.g., CL, V1, V2, Q) were evaluated. However, incorporating covariance terms did not significantly improve the model fit, as assessed by changes in OFV) and thus were not retained in the final model structures. Model refinement and selection processes were directed by variations in NONMEM OFVs, accuracy of parameter estimates indicated by relative standard errors, goodness-of-fit plots, VPC, and bootstrap analyses. Improvements in model structures were validated by decreases in OFV greater than 3.84 (for a single degree of freedom χ2 test) or 5.99 (for two degrees of freedom), which was considered statistically significant at p < 0.05. Model validation was conducted using four types of goodness-of-fit plots, including CWRES versus time, CWRES versus population predictions (PRED), observed concentrations versus PRED, and observed concentrations versus individual predictions (IPRED).VPC demonstrated that the observed concentrations of piperacillin/tazobactam were within 80% of the predicted intervals from 1000 model simulations. The variability in the final model’s predictions was examined through the medians and 95% confidence intervals derived from 2000 bootstrap samples. Significant covariates affecting PK parameters were determined through a stepwise forward selection and backward elimination process, with thresholds for inclusion at p < 0.01 (ΔOFV < −6.635) and exclusion at p < 0.001 (ΔOFV < 10.83). The covariate analysis included demographic factors such as gender, age, height, weight, LBM, body mass index, and body surface area, as well as biochemical parameters like serum protein, albumin, creatinine, and cystatin C levels. The impact of renal function on the elimination of piperacillin/tazobactam was assessed using the Cockcroft-Gault [38], Modification of Diet in Renal Disease [39], and Chronic Kidney Disease Epidemiology [15] equations. The processes of covariate identification, VPC implementation, and nonparametric bootstrapping for model stability assessment were performed using the Perl-speaks-NONMEM tool (version 5.3.1, https://uupharmacometrics.github.io/PsN/, accessed on 13 April 2023). Post-analysis processing and graphical representations were performed using the R programming environment (version 4.4.0, https://www.r-project.org/, accessed on 6 June 2024).
4.4. Dosage Simulation
To evaluate the appropriateness of current FDA and EMA dosing recommendations, a Monte Carlo simulation was conducted using the population PK model previously developed from data obtained from healthy adult subjects. A total of 20,000 virtual subjects (10,000 female, 10,000 male) were generated, whose demographic characteristics (serum creatinine, age, height, weight, and LBM) were simulated from normal distributions representative of the general adult population. Renal function for each virtual subject was calculated as the eGFR using the CKD-EPI equation, adjusted to individual body surface area (BSA). These simulated subjects were assumed to possess normal renal function. Each virtual subject received piperacillin/tazobactam at a fixed dose of 4 g/0.5 g intravenously q6h, with infusion durations of 0.5, 1, 2, 3, or 4 h. The PTA was calculated based on pathogen-specific PD targets: 50%fT>MIC for P. aeruginosa, 77%fT>MIC for E. coli, and 100%fT>MIC for K. pneumoniae. MIC distributions for these pathogens were sourced from the EUCAST database [16].
Using data from the first simulation, CFR was calculated to assess the likelihood of target attainment across various infusion strategies. The CFR represents the proportion of a simulated population expected to achieve a predefined PD target, based on MIC distributions of the relevant pathogens [11]. CFR was calculated for each pathogen by integrating the PTA across the MIC distribution, using the following formula:
where PTAi is the PTA at MIC level i, and fi is the relative frequency of that MIC value in the corresponding pathogen’s distribution. MIC distributions were obtained from the EUCAST database. CFR values were stratified by infusion duration (0.5–4 h) and compared across pathogens to assess the impact of extended infusion on population-level pharmacodynamic outcomes.
In a second simulation study, targeted dosing regimens were evaluated across five distinct renal function categories (eGFR: 0–20, 20–40, 40–90, 90–130, and 130–180 mL/min) in 1000 additional virtual subjects. Piperacillin clearance was calculated using the population PK model. The simulation explored intermittent infusion regimens (doses: 2 g, 4 g, 6 g; intervals: 6 h, 8 h, 12 h; infusion durations: 0.5 h and 3 h) and continuous infusion regimens (daily doses ranging from 4 g to 24 g). Pharmacodynamic targets evaluated were 50%fT>MIC, 50%fT>4MIC, 100%fT>MIC, and 100%fT>4MIC across MIC values from 0.5 to 128 mg/L. To specifically evaluate the impact of renal function on dosing, LBM and total body weight were fixed at their median values derived from the original study population.
Author Contributions
Conceptualization, Y.J.L. and D.H.L.; methodology, Y.J.L., G.K. and D.H.L.; software, D.H.L.; validation, G.K. and D.H.L.; formal analysis, Y.J.L., G.K. and D.H.L.; investigation, D.H.L.; resources, D.Y.Z.; data curation, Y.J.L., G.K. and D.H.L.; writing—original draft preparation, Y.J.L., G.K. and D.H.L.; writing—review and editing, Y.J.L., G.K., D.Y.Z. and D.H.L.; visualization, D.H.L.; supervision, D.H.L.; project administration, D.H.L.; funding acquisition, D.H.L. All authors have read and agreed to the published version of the manuscript.
Funding
This research was supported by the First Research in Lifetime R&D Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT (2020R1F1A1077190).
Institutional Review Board Statement
The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of the Hallym University Sacred Heart Hospital (IRB No. 2022-08-006 and date of approval 25 October 2022).
Informed Consent Statement
Written informed consent was obtained from all subjects participating in the study, including consent for publication.
Data Availability Statement
In accordance with institutional policies and to ensure participant confidentiality, the datasets generated and/or analyzed in this study cannot be shared.
Acknowledgments
We express our sincere appreciation to Young Soon Shim for her expert execution of the levofloxacin assays at the Chonnam National University Clinical Trial Center.
Conflicts of Interest
The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
Abbreviations
The following abbreviations are used in this manuscript:
| ARC | augmented renal clearance |
| BSA | body surface area |
| CC | cystatin C |
| CFR | cumulative fraction of response |
| CI | confidence interval |
| CKD-EPI | chronic kidney disease epidemiology collaboration |
| CL | total clearance |
| CR | creatinine |
| CrCl | creatinine clearance |
| CWRES | conditional weighted residuals |
| EDTA | ethylenediaminetetraacetic acid |
| eGFR | estimated glomerular filtration rate |
| EMA | European Medicines Agency |
| EUCAST | European Committee on Antimicrobial Susceptibility Testing |
| FDA | United States Food and Drug Administration |
| fT>MIC | total percentage of a 24 h period during which the concentration of the free (unbound to protein) drug surpasses the minimum inhibitory concentration under steady-state conditions |
| IIV | interindividual variability |
| IPRED | individual predictions |
| IRB | institutional review board |
| LBM | lean body mass |
| LC-MS/MS | liquid chromatography-tandem mass spectrometry |
| MDRD | modification of diet in renal disease |
| MIC | minimum inhibitory concentration |
| OFV | objective function value |
| PK/PD | pharmacokinetic/pharmacodynamic |
| PRED | population predictions |
| PTA | probability of target attainment |
| Q | intercompartmental clearance between V1 and V2 |
| RSE | relative standard error |
| V1 | volume of distribution for the central compartment |
| V2 | volume of distribution for the peripheral compartments |
| Vd | volume of distribution |
| VPC | visual predictive check |
| Vss | steady-state volume of distribution |
| WT | weight |
Appendix A
Figure A1.
Individual fit plots for piperacillin (a) normal scale, (b) semi-log scale: closed circles, observed concentrations; solid line, individual predicted concentrations; dotted line, population predicted concentrations. The numbers in each panel correspond to the individual subject number.
Figure A1.
Individual fit plots for piperacillin (a) normal scale, (b) semi-log scale: closed circles, observed concentrations; solid line, individual predicted concentrations; dotted line, population predicted concentrations. The numbers in each panel correspond to the individual subject number.
Figure A2.
Individual fit plots for tazobactam (a) normal scale, (b) semi-log scale: closed circles, observed concentrations; solid line, individual predicted concentrations; dotted line, population predicted concentrations. The numbers in each panel correspond to the individual subject number.
Figure A2.
Individual fit plots for tazobactam (a) normal scale, (b) semi-log scale: closed circles, observed concentrations; solid line, individual predicted concentrations; dotted line, population predicted concentrations. The numbers in each panel correspond to the individual subject number.
Figure A3.
Visual predictive check from simulated concentrations of 1000 virtual datasets of piperacillin (a) and tazobactam (b): closed circles, observed concentrations; solid lines, 10th, 50th, and 90th percentiles of observations; dashed lines, 10th, 50th, and 90th percentiles of simulated concentrations. The shaded areas represent the 95% confidence intervals for the simulated 10th (darkest grey), 50th (medium grey), and 90th (lightest grey) percentiles.
Figure A3.
Visual predictive check from simulated concentrations of 1000 virtual datasets of piperacillin (a) and tazobactam (b): closed circles, observed concentrations; solid lines, 10th, 50th, and 90th percentiles of observations; dashed lines, 10th, 50th, and 90th percentiles of simulated concentrations. The shaded areas represent the 95% confidence intervals for the simulated 10th (darkest grey), 50th (medium grey), and 90th (lightest grey) percentiles.
Figure A4.
Cumulative fraction of response (%) according to infusion time across pathogens, based on the probability of achieving predefined pharmacodynamic targets: 50%fT>MIC for P. aeruginosa, 77%fT>MIC for E. coli, and 100%fT>MIC for K. pneumoniae.
Figure A4.
Cumulative fraction of response (%) according to infusion time across pathogens, based on the probability of achieving predefined pharmacodynamic targets: 50%fT>MIC for P. aeruginosa, 77%fT>MIC for E. coli, and 100%fT>MIC for K. pneumoniae.
Figure A5.
Probability of target attainment (PTA) versus MIC for targeted intermittent piperacillin/tazobactam regimens achieving 50%fT>4MIC or 100%fT>4MIC across various renal function categories (eGFR). Panels compare PTA for various dosing regimens with 0.5 h versus 3 h infusion durations. Horizontal dashed lines denote the 90% PTA threshold.
Figure A5.
Probability of target attainment (PTA) versus MIC for targeted intermittent piperacillin/tazobactam regimens achieving 50%fT>4MIC or 100%fT>4MIC across various renal function categories (eGFR). Panels compare PTA for various dosing regimens with 0.5 h versus 3 h infusion durations. Horizontal dashed lines denote the 90% PTA threshold.
Table A1.
Stepwise covariate selection process of piperacillin: forward selection (p-value = 0.01, OFV difference < −6.635, degree of freedom = 1) and backward elimination (p-value = 0.001, OFV difference < 10.83, degree of freedom = 1). For comparisons between models using the same covariate but different functional forms (e.g., power vs. exponential), a degree of freedom of 0 was assumed, and a reduction in OFV > 0 was considered statistically significant.
Table A1.
Stepwise covariate selection process of piperacillin: forward selection (p-value = 0.01, OFV difference < −6.635, degree of freedom = 1) and backward elimination (p-value = 0.001, OFV difference < 10.83, degree of freedom = 1). For comparisons between models using the same covariate but different functional forms (e.g., power vs. exponential), a degree of freedom of 0 was assumed, and a reduction in OFV > 0 was considered statistically significant.
| Parameter | Covariate | Model | Base OFV | New OFV | ΔOFV | p-Value |
|---|---|---|---|---|---|---|
| Forward step 1 | ||||||
| CL | CrCl a | Power | 174.727 | 160.469 | −14.258 | 0.000159 |
| CL | eGFR, CKD-EPIMDRD b | Power | 174.727 | 161.245 | −13.482 | 0.000241 |
| CL | eGFR, CKD-EPICR c | Power | 174.727 | 158.832 | −15.895 | 0.000067 |
| CL | eGFR, CKD-EPICC d | Power | 174.727 | 163.464 | −11.263 | 0.000791 |
| CL | eGFR, CKD-EPICR-CC e | Power | 174.727 | 160.766 | −13.961 | 0.000187 |
| V2 | Body mass index | Power | 174.727 | 160.808 | −13.919 | 0.000191 |
| V2 | Body surface area | Power | 174.727 | 159.964 | −14.763 | 0.000122 |
| V2 | Height | Power | 174.727 | 164.425 | −10.302 | 0.001329 |
| V2 | Weight | Power | 174.727 | 164.619 | −10.108 | 0.001476 |
| V2 | Body mass index | Power | 174.727 | 161.257 | −13.470 | 0.000242 |
| V2 | Weight | Power | 174.727 | 159.464 | −15.263 | 0.000094 |
| Forward step 2 | ||||||
| V2 | Body mass index | Power | 158.832 | 145.594 | −13.238 | 0.000274 |
| V2 | Body surface area | Power | 158.832 | 144.621 | −14.211 | 0.000163 |
| V2 | Height | Power | 158.832 | 149.630 | −9.202 | 0.002418 |
| V2 | Weight | Power | 158.832 | 150.497 | −8.335 | 0.003890 |
| V2 | Lean body mass | Power | 158.832 | 146.621 | −12.211 | 0.000475 |
| V2 | Weight | Power | 158.832 | 144.107 | −14.725 | 0.000124 |
| Forward step 3 | ||||||
| Q | Body surface area | Power | 144.107 | 131.553 | −12.554 | 0.000395 |
| Q | Lean body mass | Power | 144.107 | 128.937 | −15.170 | 0.000098 |
| Q | Weight | Power | 144.107 | 132.535 | −11.572 | 0.000670 |
| V2 | Weight | Exponential | 144.107 | 143.488 | −0.619 | 0.000000 |
| Forward step 4 | ||||||
| Q | Body surface area | Power | 143.488 | 130.285 | −13.203 | 0.000280 |
| Q | Lean body mass | Power | 143.488 | 128.690 | −14.798 | 0.000120 |
| Q | Weight | Power | 143.488 | 130.687 | −12.801 | 0.000346 |
| Backward step 1 | ||||||
| CL | eGFR, CKD-EPICR c | None | 128.690 | 145.105 | 16.414 | 0.000051 |
| Q | Lean body mass | None | 128.690 | 143.488 | 14.798 | 0.000120 |
| V2 | Weight | Power | 128.690 | 128.937 | 0.246 | 0.000000 |
OFV, objective function value; ΔOFV = New OFV − Base OFV; CrCl, creatinine clearance by Cockcroft-Gault equation; eGFR, estimated glomerular filtration rate; MDRD, modification of diet in renal disease; CKD-EPI, chronic kidney disease epidemiology collaboration; CR, creatinine; CC, cystatin C; min, the minimum of (CR or CC)/number and 1; max, the maximum of (CR or CC)/number and 1. a CrCl = (140 − Age) × weight/CR × 72 (×0.85 if female); b eGFR = 175 × CR−1.154 × Age−0.203 (×0.742 if female); c eGFR (female) = 142 × min (CR/0.7,1)−0.241 × max (CR/0.7,1)−1.200 × 0.9938Age × 1.012/1.73 × BSA; eGFR (male) = 142 × min (CR/0.9,1)−0.302 × max (CR/0.9,1)−1.200 × 0.9938 Age/1.73 × BSA; d eGFR = 133 × min (CC/0.8,1)−0.499 × max (CC/0.8,1)−1.328 × 0.9962Age × 0.932 [if female]/1.73 × BSA; e eGFR (female) = 135 × min (CR/0.7,1)−0.219 × max (CR/0.7,1)−0.544 × min (CC/0.8,1)0.323 × max (CC/0.8,1)−0.778 × 0.9961Age × 0.963 /1.73 × BSA.; eGFR (male) = 135 × min (CR/0.9,1)−0.144 × max (CR/0.9,1)−0.544 × min (CC/0.8,1)0.323 × max (CC/0.8,1)−0.778 × 0.9961Age/1.73 × BSA. Values in red and bold indicate the covariate that was selected in each step of the stepwise analysis.
Table A2.
Stepwise covariate selection process of tazobactam: forward selection (p-value = 0.01, OFV difference < −6.635, degree of freedom = 1) and backward elimination (p-value = 0.001, OFV difference < 10.83, degree of freedom = 1). For comparisons between models using the same covariate but different functional forms (e.g., power vs. exponential), a degree of freedom of 0 was assumed, and a reduction in OFV > 0 was considered statistically significant.
Table A2.
Stepwise covariate selection process of tazobactam: forward selection (p-value = 0.01, OFV difference < −6.635, degree of freedom = 1) and backward elimination (p-value = 0.001, OFV difference < 10.83, degree of freedom = 1). For comparisons between models using the same covariate but different functional forms (e.g., power vs. exponential), a degree of freedom of 0 was assumed, and a reduction in OFV > 0 was considered statistically significant.
| Parameter | Covariate | Model | Base OFV | New OFV | ΔOFV | p-Value |
|---|---|---|---|---|---|---|
| Forward step 1 | ||||||
| CL | CrCl a | Power | −157.083 | −168.790 | −11.707 | 0.000623 |
| CL | eGFR, CKD-EPIMDRD b | Power | −157.083 | −167.523 | −10.440 | 0.001233 |
| CL | eGFR, CKD-EPICR c | Power | −157.083 | −168.795 | −11.712 | 0.000621 |
| CL | eGFR, CKD-EPICC d | Power | −157.083 | −165.252 | −8.169 | 0.004260 |
| CL | eGFR, CKD-EPICR-CC e | Power | −157.083 | −166.989 | −9.906 | 0.001647 |
| CL | Weight | Power | −157.083 | −164.717 | −7.634 | 0.005727 |
| CL | Body surface area | Power | −157.083 | −164.675 | −7.592 | 0.005861 |
| V1 | Height | Power | −157.083 | −164.183 | −7.100 | 0.007710 |
| V2 | Body mass index | Power | −157.083 | −167.052 | −9.969 | 0.001592 |
| V2 | Body surface area | Power | −157.083 | −167.924 | −10.841 | 0.000993 |
| V2 | Lean body mass | Power | −157.083 | −168.305 | −11.222 | 0.000808 |
| V2 | Weight | Power | −157.083 | −168.610 | −11.527 | 0.000686 |
| Forward step 2 | ||||||
| V1 | Height | Power | −168.795 | −176.628 | −7.833 | 0.005129 |
| V2 | Body mass index | Power | −168.795 | −179.349 | −10.554 | 0.001159 |
| V2 | Body surface area | Power | −168.795 | −181.942 | −13.147 | 0.000288 |
| V2 | Height | Power | −168.795 | −176.106 | −7.312 | 0.006850 |
| V2 | Lean body mass | Power | −168.795 | −181.863 | −13.068 | 0.000300 |
| V2 | Weight | Power | −168.795 | −182.092 | −13.297 | 0.000266 |
| Forward step 3 | ||||||
| Q | Body surface area | Power | −182.092 | −192.224 | −10.132 | 0.001457 |
| Q | Lean body mass | Power | −182.092 | −191.539 | −9.447 | 0.002115 |
| Q | Weight | Power | −182.092 | −191.719 | −9.627 | 0.001917 |
| V2 | Weight | Exponential | −182.092 | −182.631 | −0.539 | 0.000000 |
| Forward step 4 | ||||||
| Q | Body surface area | Power | −182.631 | −192.957 | −10.326 | 0.001312 |
| Q | Lean body mass | Power | −182.631 | −191.840 | −9.209 | 0.002409 |
| Q | Weight | Power | −182.631 | −192.795 | −10.164 | 0.001432 |
| V1 | Height | Power | −182.631 | −189.746 | −7.115 | 0.007645 |
| Backward step 1 | ||||||
| CL | eGFR, CKD-EPICR c | None | −192.957 | −179.223 | 13.733 | 0.000211 |
| Q | Body surface area | None | −192.957 | −182.631 | 10.326 | 0.001312 |
| V2 | Weight | Power | −192.957 | −192.224 | 0.733 | 0.000000 |
| Backward step 2 | ||||||
| CL | eGFR, CKD-EPICR c | None | −182.631 | −169.313 | 13.318 | 0.000263 |
| V2 | Weight | Power | −182.631 | −182.092 | 0.539 | 0.000000 |
OFV, objective function value; ΔOFV = New OFV − Base OFV; CrCl, creatinine clearance by Cockcroft-Gault equation; eGFR, estimated glomerular filtration rate; MDRD, modification of diet in renal disease; CKD-EPI, chronic kidney disease epidemiology collaboration; CR, creatinine; CC, cystatin C; min, the minimum of (CR or CC)/number and 1; max, the maximum of (CR or CC)/number and 1. a CrCl = (140 − Age) × weight/CR × 72 (×0.85 if female); b eGFR = 175 × CR−1.154 × Age−0.203 (×0.742 if female); c eGFR (female) = 142 × min (CR/0.7,1)−0.241 × max (CR/0.7,1)−1.200 × 0.9938Age × 1.012/1.73 × BSA; eGFR (male) = 142 × min (CR/0.9,1)−0.302 × max (CR/0.9,1)−1.200 × 0.9938 Age/1.73 × BSA; d eGFR = 133 × min (CC/0.8,1)−0.499 × max (CC/0.8,1)−1.328 × 0.9962Age × 0.932 [if female]/1.73 × BSA; e eGFR (female) = 135 × min (CR/0.7,1)−0.219 × max (CR/0.7,1)−0.544 × min (CC/0.8,1)0.323 × max (CC/0.8,1)−0.778 × 0.9961Age × 0.963 /1.73 × BSA; eGFR (male) = 135 × min (CR/0.9,1)−0.144 × max (CR/0.9,1)−0.544 × min (CC /0.8,1)0.323 × max (CC /0.8,1)−0.778 × 0.9961Age/1.73 × BSA. Values in red and bold indicate the covariate that was selected in each step of the stepwise analysis.
Table A3.
Recommended intermittent piperacillin/tazobactam dosing regimens (probability of target attainment, PTA) to achieve ≥90% PTA for the 50%fT>MIC or 100%fT>MIC targets, stratified by renal function category (eGFR) and MIC value.
Table A3.
Recommended intermittent piperacillin/tazobactam dosing regimens (probability of target attainment, PTA) to achieve ≥90% PTA for the 50%fT>MIC or 100%fT>MIC targets, stratified by renal function category (eGFR) and MIC value.
| MIC | Target: 50%fT>MIC | Target: 100%fT>MIC | |||
|---|---|---|---|---|---|
| 0.5 h Infusion | 3 h Infusion | 0.5 h Infusion | 3 h Infusion | CI | |
| BSA adjusted eGFR: 0–20 mL/min | |||||
| 0.5 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 1 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 2 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 4 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 8 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 16 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (98.3) | 2 g q12h (100) | 4 g (100) |
| 32 | 2 g q12h (100) | 2 g q12h (100) | 2 g q8h (100) | 2 g q12h (91.5) | 4 g (100) |
| 64 | 2 g q12h (94.1) | 2 g q12h (96.6) | 2 g q6h (100) | 2 g q8h (94.1) | 4 g (100) |
| 128 | 2 g q6h (96.6) | 2 g q6h (100) | 4 g q6h (100) | 4 g q8h (94.1) | 8 g (100) |
| BSA adjusted eGFR: 20–40 mL/min | |||||
| 0.5 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 1 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 2 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (94.6) | 2 g q12h (100) | 4 g (100) |
| 4 | 2 g q12h (100) | 2 g q12h (100) | 2 g q8h (100) | 2 g q8h (100) | 4 g (100) |
| 8 | 2 g q12h (100) | 2 g q12h (100) | 2 g q8h (96.8) | 2 g q8h (100) | 4 g (100) |
| 16 | 2 g q12h (98.9) | 2 g q12h (100) | 2 g q6h (100) | 2 g q6h (100) | 4 g (100) |
| 32 | 2 g q8h (100) | 2 g q8h (100) | 4 g q6h (100) | 2 g q6h (95.7) | 4 g (100) |
| 64 | 4 g q8h (100) | 2 g q6h (100) | 6 g q6h (89.2) | 4 g q6h (95.7) | 8 g (100) |
| 128 | 6 g q6h (100) | 4 g q6h (100) | 6 g q6h (40.9) | 6 g q6h (66.7) | 16 g (100) |
| BSA adjusted eGFR: 40–90 mL/min | |||||
| 0.5 | 2 g q12h (99.3) | 2 g q12h (100) | 2 g q6h (99.3) | 2 g q8h (97.2) | 4 g (100) |
| 1 | 2 g q12h (91.5) | 2 g q12h (100) | 2 g q6h (91.5) | 2 g q6h (100) | 4 g (100) |
| 2 | 2 g q8h (100) | 2 g q12h (99.3) | 4 g q6h (91.5) | 2 g q6h (99.3) | 4 g (100) |
| 4 | 2 g q8h (96.8) | 2 g q8h (100) | 6 g q6h (83.1) | 4 g q6h (99.3) | 4 g (100) |
| 8 | 2 g q6h (98.9) | 2 g q8h (100) | 6 g q6h (64.8) | 6 g q6h (97.2) | 4 g (100) |
| 16 | 4 g q6h (98.9) | 2 g q6h (100) | 6 g q6h (44.0) | 6 g q6h (75.4) | 8 g (100) |
| 32 | 6 g q6h (91.2) | 4 g q6h (100) | 6 g q6h (20.1) | 6 g q6h (46.1) | 12 g (100) |
| 64 | 6 g q6h (54.6) | 6 g q6h (100) | – | 6 g q6h (17.6) | 20 g (96.1) |
| 128 | 6 g q6h (14.8) | 6 g q6h (42.6) | – | – | 24 g (42.6) |
| BSA adjusted eGFR: 90–130 mL/min | |||||
| 0.5 | 2 g q8h (100) | 2 g q12h (98.7) | 6 g q6h (83.6) | 2 g q6h (98.7) | 4 g (100) |
| 1 | 2 g q8h (93.5) | 2 g q8h (100) | 6 g q6h (45.7) | 4 g q6h (98.7) | 4 g (100) |
| 2 | 2 g q6h (100) | 2 g q8h (100) | 6 g q6h (19.0) | 6 g q6h (93.5) | 4 g (100) |
| 4 | 4 g q6h (100) | 2 g q8h (100) | – | 6 g q6h (53.0) | 4 g (100) |
| 8 | 6 g q6h (96.1) | 2 g q6h (100) | – | 6 g q6h (15.5) | 4 g (98.7) |
| 16 | 6 g q6h (51.3) | 2 g q6h (98.7) | – | – | 8 g (98.7) |
| 32 | – | 4 g q6h (98.7) | – | – | 16 g (98.7) |
| 64 | – | 6 g q6h (40.9) | – | – | 24 g (39.2) |
| 128 | – | – | – | – | – |
| BSA adjusted eGFR: 130–180 mL/min | |||||
| 0.5 | 2 g q6h (100) | 2 g q8h (100) | – | 6 g q6h (93.4) | 4 g (100) |
| 1 | 4 g q6h (100) | 2 g q8h (100) | – | 6 g q6h (52.7) | 4 g (100) |
| 2 | 6 g q6h (99.6) | 2 g q8h (96.3) | – | 6 g q6h (14.3) | 4 g (100) |
| 4 | 6 g q6h (69.6) | 2 g q6h (100) | – | – | 4 g (100) |
| 8 | 6 g q6h (20.1) | 2 g q6h (100) | – | – | 8 g (100) |
| 16 | – | 4 g q6h (100) | – | – | 12 g (98.2) |
| 32 | – | 6 g q6h (98.2) | – | – | 24 g (98.2) |
| 64 | – | – | – | – | – |
| 128 | – | – | – | – | – |
CI, continuous infusion; BSA, body surface area. Shaded cells indicate dosing regimens that achieved ≥90% PTA.
Table A4.
Recommended intermittent piperacillin/tazobactam dosing regimens (probability of target attainment, PTA) to achieve ≥90% PTA for the 50%fT4>MIC or 100%fT>4MIC targets, stratified by renal function category (eGFR) and MIC value.
Table A4.
Recommended intermittent piperacillin/tazobactam dosing regimens (probability of target attainment, PTA) to achieve ≥90% PTA for the 50%fT4>MIC or 100%fT>4MIC targets, stratified by renal function category (eGFR) and MIC value.
| MIC | Target: 50%fT>4MIC | Target: 100%fT>4MIC | |||
|---|---|---|---|---|---|
| 0.5 h Infusion | 3 h Infusion | 0.5 h Infusion | 3 h Infusion | CI | |
| BSA adjusted eGFR: 0–20 mL/min | |||||
| 0.5 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 1 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 2 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (100) | 4 g (100) |
| 4 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (98.3) | 2 g q12h (100) | 4 g (100) |
| 8 | 2 g q12h (100) | 2 g q12h (100) | 2 g q8h (100) | 2 g q12h (91.5) | 4 g (100) |
| 16 | 2 g q12h (94.1) | 2 g q12h (96.6) | 2 g q6h (100) | 2 g q8h (94.1) | 4 g (100) |
| 32 | 2 g q6h (96.6) | 2 g q6h (100) | 4 g q6h (100) | 4 g q8h (94.1) | 8 g (100) |
| 64 | 4 g q6h (96.6) | 4 g q6h (100) | 6 g q6h (94.1) | 6 g q6h (100) | 16 g (100) |
| 128 | 6 g q6h (80.5) | 6 g q6h (88.1) | 6 g q6h (63.6) | 6 g q6h (71.2) | 24 g (88.1) |
| BSA adjusted eGFR: 20–40 mL/min | |||||
| 0.5 | 2 g q12h (100) | 2 g q12h (100) | 2 g q12h (94.6) | 2 g q12h (100) | 4 g (100) |
| 1 | 2 g q12h (100) | 2 g q12h (100) | 2 g q8h (100) | 2 g q8h (100) | 4 g (100) |
| 2 | 2 g q12h (100) | 2 g q12h (100) | 2 g q8h (96.8) | 2 g q8h (100) | 4 g (100) |
| 4 | 2 g q12h (98.9) | 2 g q12h (100) | 2 g q6h (100) | 2 g q6h (100) | 4 g (100) |
| 8 | 2 g q8h (100) | 2 g q8h (100) | 4 g q6h (100) | 2 g q6h (95.7) | 4 g (100) |
| 16 | 4 g q8h (100) | 2 g q6h (100) | 6 g q6h (89.2) | 4 g q6h (95.7) | 8 g (100) |
| 32 | 6 g q6h (100) | 4 g q6h (100) | 6 g q6h (40.9) | 6 g q6h (66.7) | 16 g (100) |
| 64 | 6 g q6h (44.1) | 6 g q6h (58.1) | – | 6 g q6h (23.7) | 24 g (58.1) |
| 128 | – | – | – | – | – |
| BSA adjusted eGFR: 40–90 mL/min | |||||
| 0.5 | 2 g q8h (100) | 2 g q12h (99.3) | 4 g q6h (91.5) | 2 g q6h (99.3) | 4 g (100) |
| 1 | 2 g q8h (96.8) | 2 g q8h (100) | 6 g q6h (83.1) | 4 g q6h (99.3) | 4 g (100) |
| 2 | 2 g q6h (98.9) | 2 g q8h (100) | 6 g q6h (64.8) | 6 g q6h (97.2) | 4 g (100) |
| 4 | 4 g q6h (98.9) | 2 g q6h (100) | 6 g q6h (44.0) | 6 g q6h (75.4) | 8 g (100) |
| 8 | 6 g q6h (91.2) | 4 g q6h (100) | 6 g q6h (20.1) | 6 g q6h (46.1) | 12 g (100) |
| 16 | 6 g q6h (54.6) | 6 g q6h (100) | – | 6 g q6h (17.6) | 20 g (96.1) |
| 32 | 6 g q6h (14.8) | 6 g q6h (42.6) | – | – | 24 g (42.6) |
| 64 | – | – | – | – | – |
| 128 | – | – | – | – | – |
| BSA adjusted eGFR: 90–130 mL/min | |||||
| 0.5 | 2 g q6h (100) | 2 g q8h (100) | 6 g q6h (19) | 6 g q6h (93.5) | 4 g (100) |
| 1 | 4 g q6h (100) | 2 g q8h (100) | – | 6 g q6h (53.0) | 4 g (100) |
| 2 | 6 g q6h (96.1) | 2 g q6h (100) | – | 6 g q6h (15.5) | 4 g (98.7) |
| 4 | 6 g q6h (51.3) | 2 g q6h (98.7) | – | – | 8 g (98.7) |
| 8 | – | 4 g q6h (98.7) | – | – | 16 g (98.7) |
| 16 | – | 6 g q6h (40.9) | – | – | 24 g (39.2) |
| 32 | – | – | – | – | – |
| 64 | – | – | – | – | – |
| 128 | – | – | – | – | – |
| BSA adjusted eGFR: 130–180 mL/min | |||||
| 0.5 | 6 g q6h (99.6) | 2 g q8h (96.3) | – | 6 g q6h (14.3) | 4 g (100) |
| 1 | 6 g q6h (69.6) | 2 g q6h (100) | – | – | 4 g (100) |
| 2 | 6 g q6h (20.1) | 2 g q6h (100) | – | – | 8 g (100) |
| 4 | – | 4 g q6h (100) | – | – | 12 g (98.2) |
| 8 | – | 6 g q6h (98.2) | – | – | 24 g (98.2) |
| 16 | – | – | – | – | – |
| 32 | – | – | – | – | – |
| 64 | – | – | – | – | – |
| 128 | – | – | – | – | – |
CI, continuous infusion; BSA, body surface area. Shaded cells indicate dosing regimens that achieved ≥90% PTA.
References
- U.S. Food and Drug Administration. Zosyn. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=050750 (accessed on 7 July 2024).
- European Medicines Agency. Tazocin. Available online: https://www.ema.europa.eu/en/medicines/human/referrals/tazocin (accessed on 10 August 2024).
- Ng, T.M.; Khong, W.X.; Harris, P.N.; De, P.P.; Chow, A.; Tambyah, P.A.; Lye, D.C. Empiric Piperacillin-Tazobactam versus Carbapenems in the Treatment of Bacteraemia Due to Extended-Spectrum Beta-Lactamase-Producing Enterobacteriaceae. PLoS ONE 2016, 11, e0153696. [Google Scholar] [CrossRef]
- Babich, T.; Naucler, P.; Valik, J.K.; Giske, C.G.; Benito, N.; Cardona, R.; Rivera, A.; Pulcini, C.; Abdel Fattah, M.; Haquin, J.; et al. Ceftazidime, Carbapenems, or Piperacillin-tazobactam as Single Definitive Therapy for Pseudomonas aeruginosa Bloodstream Infection: A Multisite Retrospective Study. Clin. Infect. Dis. 2020, 70, 2270–2280. [Google Scholar] [CrossRef]
- De Bus, L.; Depuydt, P.; Steen, J.; Dhaese, S.; De Smet, K.; Tabah, A.; Akova, M.; Cotta, M.O.; De Pascale, G.; Dimopoulos, G.; et al. Antimicrobial de-escalation in the critically ill patient and assessment of clinical cure: The DIANA study. Intensive Care Med. 2020, 46, 1404–1417. [Google Scholar] [CrossRef]
- Bryson, H.M.; Brogden, R.N. Piperacillin/tazobactam. A review of its antibacterial activity, pharmacokinetic properties and therapeutic potential. Drugs 1994, 47, 506–535. [Google Scholar] [CrossRef]
- Johnson, C.A.; Halstenson, C.E.; Kelloway, J.S.; Shapiro, B.E.; Zimmerman, S.W.; Tonelli, A.; Faulkner, R.; Dutta, A.; Haynes, J.; Greene, D.S.; et al. Single-dose pharmacokinetics of piperacillin and tazobactam in patients with renal disease. Clin. Pharmacol. Ther. 1992, 51, 32–41. [Google Scholar] [CrossRef]
- Roberts, J.A.; Lipman, J. Pharmacokinetic issues for antibiotics in the critically ill patient. Crit. Care Med. 2009, 37, 840–851. [Google Scholar] [CrossRef]
- Craig, W.A. Pharmacokinetic/pharmacodynamic parameters: Rationale for antibacterial dosing of mice and men. Clin. Infect. Dis. 1998, 26, 1–12. [Google Scholar] [CrossRef]
- Drusano, G.L. Antimicrobial pharmacodynamics: Critical interactions of ‘bug and drug’. Nat. Rev. Microbiol. 2004, 2, 289–300. [Google Scholar] [CrossRef]
- Mouton, J.W.; Dudley, M.N.; Cars, O.; Derendorf, H.; Drusano, G.L. Standardization of pharmacokinetic/pharmacodynamic (PK/PD) terminology for anti-infective drugs: An update. J. Antimicrob. Chemother. 2005, 55, 601–607. [Google Scholar] [CrossRef]
- Asin-Prieto, E.; Rodriguez-Gascon, A.; Troconiz, I.F.; Soraluce, A.; Maynar, J.; Sanchez-Izquierdo, J.A.; Isla, A. Population pharmacokinetics of piperacillin and tazobactam in critically ill patients undergoing continuous renal replacement therapy: Application to pharmacokinetic/pharmacodynamic analysis. J. Antimicrob. Chemother. 2014, 69, 180–189. [Google Scholar] [CrossRef]
- Sime, F.B.; Hahn, U.; Warner, M.S.; Tiong, I.S.; Roberts, M.S.; Lipman, J.; Peake, S.L.; Roberts, J.A. Using Population Pharmacokinetic Modeling and Monte Carlo Simulations To Determine whether Standard Doses of Piperacillin in Piperacillin-Tazobactam Regimens Are Adequate for the Management of Febrile Neutropenia. Antimicrob. Agents Chemother. 2017, 61, 10.1128. [Google Scholar] [CrossRef]
- Ishihara, N.; Nishimura, N.; Ikawa, K.; Karino, F.; Miura, K.; Tamaki, H.; Yano, T.; Isobe, T.; Morikawa, N.; Naora, K. Population Pharmacokinetic Modeling and Pharmacodynamic Target Attainment Simulation of Piperacillin/Tazobactam for Dosing Optimization in Late Elderly Patients with Pneumonia. Antibiotics 2020, 9, 113. [Google Scholar] [CrossRef]
- Inker, L.A.; Eneanya, N.D.; Coresh, J.; Tighiouart, H.; Wang, D.; Sang, Y.; Crews, D.C.; Doria, A.; Estrella, M.M.; Froissart, M.; et al. New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race. N. Engl. J. Med. 2021, 385, 1737–1749. [Google Scholar] [CrossRef]
- European Committee on Antimicrobial Susceptibility Testing. Antimicrobial Wild Type Distributions of Microorganisms. Available online: https://mic.eucast.org/ (accessed on 27 March 2024).
- Gonzalez-Sales, M.; Holford, N.; Bonnefois, G.; Desrochers, J. Wide size dispersion and use of body composition and maturation improves the reliability of allometric exponent estimates. J. Pharmacokinet. Pharmacodyn. 2022, 49, 151–165. [Google Scholar] [CrossRef]
- Daniel, K.P.; Krop, L.C. Piperacillin-Tazobactam: A new beta-lactam-beta-lactamase inhibitor combination. Pharmacotherapy 1996, 16, 149–162. [Google Scholar] [CrossRef]
- Bulitta, J.B.; Duffull, S.B.; Kinzig-Schippers, M.; Holzgrabe, U.; Stephan, U.; Drusano, G.L.; Sorgel, F. Systematic comparison of the population pharmacokinetics and pharmacodynamics of piperacillin in cystic fibrosis patients and healthy volunteers. Antimicrob. Agents Chemother. 2007, 51, 2497–2507. [Google Scholar] [CrossRef]
- Bulitta, J.B.; Kinzig, M.; Jakob, V.; Holzgrabe, U.; Sorgel, F.; Holford, N.H. Nonlinear pharmacokinetics of piperacillin in healthy volunteers--implications for optimal dosage regimens. Br. J. Clin. Pharmacol. 2010, 70, 682–693. [Google Scholar] [CrossRef]
- Felton, T.W.; Ogungbenro, K.; Boselli, E.; Hope, W.W.; Rodvold, K.A. Comparison of piperacillin exposure in the lungs of critically ill patients and healthy volunteers. J. Antimicrob. Chemother. 2018, 73, 1340–1347. [Google Scholar] [CrossRef]
- Roberts, J.A.; Kirkpatrick, C.M.; Roberts, M.S.; Dalley, A.J.; Lipman, J. First-dose and steady-state population pharmacokinetics and pharmacodynamics of piperacillin by continuous or intermittent dosing in critically ill patients with sepsis. Int. J. Antimicrob. Agents 2010, 35, 156–163. [Google Scholar] [CrossRef]
- Udy, A.A.; Lipman, J.; Jarrett, P.; Klein, K.; Wallis, S.C.; Patel, K.; Kirkpatrick, C.M.; Kruger, P.S.; Paterson, D.L.; Roberts, M.S.; et al. Are standard doses of piperacillin sufficient for critically ill patients with augmented creatinine clearance? Crit. Care 2015, 19, 28. [Google Scholar] [CrossRef]
- Kim, Y.K.; Kim, H.S.; Park, S.; Kim, H.I.; Lee, S.H.; Lee, D.H. Population pharmacokinetics of piperacillin/tazobactam in critically ill Korean patients and the effects of extracorporeal membrane oxygenation. J. Antimicrob. Chemother. 2022, 77, 1353–1364. [Google Scholar] [CrossRef]
- Chen, R.; Qian, Q.; Sun, M.R.; Qian, C.Y.; Zou, S.L.; Wang, M.L.; Wang, L.Y. Population Pharmacokinetics and Pharmacodynamics of Piperacillin/Tazobactam in Patients with Nosocomial Infections. Eur. J. Drug Metab. Pharmacokinet. 2016, 41, 363–372. [Google Scholar] [CrossRef]
- Patel, N.; Scheetz, M.H.; Drusano, G.L.; Lodise, T.P. Identification of optimal renal dosage adjustments for traditional and extended-infusion piperacillin-tazobactam dosing regimens in hospitalized patients. Antimicrob. Agents Chemother. 2010, 54, 460–465. [Google Scholar] [CrossRef]
- Felton, T.W.; Hope, W.W.; Lomaestro, B.M.; Butterfield, J.M.; Kwa, A.L.; Drusano, G.L.; Lodise, T.P. Population pharmacokinetics of extended-infusion piperacillin-tazobactam in hospitalized patients with nosocomial infections. Antimicrob. Agents Chemother. 2012, 56, 4087–4094. [Google Scholar] [CrossRef]
- Cies, J.J.; Shankar, V.; Schlichting, C.; Kuti, J.L. Population pharmacokinetics of piperacillin/tazobactam in critically ill young children. Pediatr. Infect. Dis. J. 2014, 33, 168–173. [Google Scholar] [CrossRef]
- Roberts, J.A.; Paul, S.K.; Akova, M.; Bassetti, M.; De Waele, J.J.; Dimopoulos, G.; Kaukonen, K.M.; Koulenti, D.; Martin, C.; Montravers, P.; et al. DALI: Defining antibiotic levels in intensive care unit patients: Are current beta-lactam antibiotic doses sufficient for critically ill patients? Clin. Infect. Dis. 2014, 58, 1072–1083. [Google Scholar] [CrossRef]
- Beranger, A.; Benaboud, S.; Urien, S.; Moulin, F.; Bille, E.; Lesage, F.; Zheng, Y.; Genuini, M.; Gana, I.; Renolleau, S.; et al. Piperacillin Population Pharmacokinetics and Dosing Regimen Optimization in Critically Ill Children with Normal and Augmented Renal Clearance. Clin. Pharmacokinet. 2019, 58, 223–233. [Google Scholar] [CrossRef]
- Lonsdale, D.O.; Kipper, K.; Baker, E.H.; Barker, C.I.S.; Oldfield, I.; Philips, B.J.; Johnston, A.; Rhodes, A.; Sharland, M.; Standing, J.F. beta-Lactam antimicrobial pharmacokinetics and target attainment in critically ill patients aged 1 day to 90 years: The ABDose study. J. Antimicrob. Chemother. 2020, 75, 3625–3634. [Google Scholar] [CrossRef]
- Chongcharoenyanon, T.; Wacharachaisurapol, N.; Anugulruengkitt, S.; Maimongkol, P.; Anunsittichai, O.; Sophonphan, J.; Chatsuwan, T.; Puthanakit, T. Comparison of piperacillin plasma concentrations in a prospective randomised trial of extended infusion versus intermittent bolus of piperacillin/tazobactam in paediatric patients. Int. J. Infect. Dis. 2021, 108, 102–108. [Google Scholar] [CrossRef]
- Udy, A.A.; Baptista, J.P.; Lim, N.L.; Joynt, G.M.; Jarrett, P.; Wockner, L.; Boots, R.J.; Lipman, J. Augmented renal clearance in the ICU: Results of a multicenter observational study of renal function in critically ill patients with normal plasma creatinine concentrations*. Crit. Care Med. 2014, 42, 520–527. [Google Scholar] [CrossRef]
- Klastrup, V.; Thorsted, A.; Storgaard, M.; Christensen, S.; Friberg, L.E.; Obrink-Hansen, K. Population Pharmacokinetics of Piperacillin following Continuous Infusion in Critically Ill Patients and Impact of Renal Function on Target Attainment. Antimicrob. Agents Chemother. 2020, 64, 10.1128. [Google Scholar] [CrossRef]
- Colman, S.; Stove, V.; De Waele, J.J.; Verstraete, A.G. Measuring Unbound Versus Total Piperacillin Concentrations in Plasma of Critically Ill Patients: Methodological Issues and Relevance. Ther. Drug Monit. 2019, 41, 325–330. [Google Scholar] [CrossRef]
- Al-Shaer, M.H.; Alghamdi, W.A.; Graham, E.; Peloquin, C.A. Meropenem, Cefepime, and Piperacillin Protein Binding in Patient Samples. Ther. Drug Monit. 2020, 42, 129–132. [Google Scholar] [CrossRef]
- El-Haffaf, I.; Guilhaumou, R.; Velly, L.; Marsot, A. Impact of piperacillin unbound fraction variability on dosing recommendations in critically ill patients. Br. J. Clin. Pharmacol. 2023, 89, 1502–1508. [Google Scholar] [CrossRef]
- Cockcroft, D.W.; Gault, M.H. Prediction of creatinine clearance from serum creatinine. Nephron 1976, 16, 31–41. [Google Scholar] [CrossRef]
- Levey, A.S.; Coresh, J.; Greene, T.; Stevens, L.A.; Zhang, Y.L.; Hendriksen, S.; Kusek, J.W.; Van Lente, F.; Chronic Kidney Disease Epidemiology, C. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann. Intern. Med. 2006, 145, 247–254. [Google Scholar] [CrossRef]
Figure 1.
Piperacillin (a) and tazobactam (b) concentration–time profile in healthy adults. The open circles denote the observed plasma concentrations.
Figure 1.
Piperacillin (a) and tazobactam (b) concentration–time profile in healthy adults. The open circles denote the observed plasma concentrations.
Figure 2.
Goodness-of-fit plots for the final pharmacokinetic model of piperacillin (upper) and tazobactam (lower): (a,e) conditional weighted residuals (CWRES) vs. time, (b,f) CWRES vs. population predicted concentration (PRED) (c,g) observed concentration vs. PRED and (d,h) observed concentration vs. individual predicted concentration (IPRED). The dashed lines represent smoothing spline curves.
Figure 2.
Goodness-of-fit plots for the final pharmacokinetic model of piperacillin (upper) and tazobactam (lower): (a,e) conditional weighted residuals (CWRES) vs. time, (b,f) CWRES vs. population predicted concentration (PRED) (c,g) observed concentration vs. PRED and (d,h) observed concentration vs. individual predicted concentration (IPRED). The dashed lines represent smoothing spline curves.
Figure 3.
Probability of target attainment (PTA) versus MIC for the empirical piperacillin/tazobactam regimen (4 g/0.5 g q6h) in virtual subjects with normal renal function. Panels show PTA for different infusion durations (0.5, 1, 2, 3, and 4 h) against (A) Pseudomonas aeruginosa (Target: 50%fT>MIC), (B) Escherichia coli (Target: 77%fT>MIC), and (C) Klebsiella pneumoniae (Target: 100%fT>MIC). Gray bars represent the MIC frequency distribution for each pathogen. Horizontal dashed lines denote the 90% PTA threshold.
Figure 3.
Probability of target attainment (PTA) versus MIC for the empirical piperacillin/tazobactam regimen (4 g/0.5 g q6h) in virtual subjects with normal renal function. Panels show PTA for different infusion durations (0.5, 1, 2, 3, and 4 h) against (A) Pseudomonas aeruginosa (Target: 50%fT>MIC), (B) Escherichia coli (Target: 77%fT>MIC), and (C) Klebsiella pneumoniae (Target: 100%fT>MIC). Gray bars represent the MIC frequency distribution for each pathogen. Horizontal dashed lines denote the 90% PTA threshold.
Figure 4.
Probability of target attainment (PTA) versus MIC for targeted intermittent piperacillin/tazobactam regimens achieving 50%fT>MIC or 100%fT>MIC across various renal function categories (eGFR). Panels compare PTA for different doses (2 g, 4 g, or 6 g) and dosing intervals (6 h, 8 h, or 12 h) with 0.5 h versus 3 h infusion durations. Horizontal dashed lines denote the 90% PTA threshold.
Figure 4.
Probability of target attainment (PTA) versus MIC for targeted intermittent piperacillin/tazobactam regimens achieving 50%fT>MIC or 100%fT>MIC across various renal function categories (eGFR). Panels compare PTA for different doses (2 g, 4 g, or 6 g) and dosing intervals (6 h, 8 h, or 12 h) with 0.5 h versus 3 h infusion durations. Horizontal dashed lines denote the 90% PTA threshold.
Figure 5.
Probability of target attainment (PTA) for continuous infusion regimens of piperacillin across different renal function categories. PTA values were evaluated for achieving pharmacodynamic targets of 100%fT>MIC or 100%fT>4MIC at various MIC levels (0.5–128 mg/L). Daily doses ranged from 4 to 24 g/day. Horizontal dashed lines denote the 90% PTA threshold.
Figure 5.
Probability of target attainment (PTA) for continuous infusion regimens of piperacillin across different renal function categories. PTA values were evaluated for achieving pharmacodynamic targets of 100%fT>MIC or 100%fT>4MIC at various MIC levels (0.5–128 mg/L). Daily doses ranged from 4 to 24 g/day. Horizontal dashed lines denote the 90% PTA threshold.
Table 1.
Characteristics of the participants.
| Parameters | Total (n = 12) | Female (n = 4) | Male (n = 8) |
|---|---|---|---|
| Demographic characteristics | |||
| Age, years | 36.0 (26.0–50.0) | 29.5 (26.0–36.0) | 39.0 (32.0–50.0) |
| Height, cm | 168 (158–182) | 163 (158–167) | 171 (160–182) |
| Weight, kg | 61.7 (45.8–88.5) | 56.1 (45.8–59.5) | 69.1 (54.5–88.5) |
| Lean body mass, kg | 50.1 (36.6–65.9) | 42.6 (36.6–44.8) | 55.2 (45.1–65.9) |
| Body surface area, m2 | 1.71 (1.44–2.07) | 1.60 (1.44–1.66) | 1.81 (1.56–2.07) |
| Body mass index, kg/m2 | 21.5 (18.3–29.7) | 21.0 (18.3–21.5) | 23.5 (21.3–29.7) |
| Laboratory characteristics | |||
| Protein, g/dL | 7.45 (7.00–8.30) | 7.45 (7.30–8.30) | 7.45 (7.00–7.70) |
| Albumin, g/dL | 4.80 (4.60–5.20) | 4.90 (4.70–5.20) | 4.80 (4.60–5.20) |
| Cystatin C, mg/dL | 0.765 (0.620–1.01) | 0.700 (0.620–0.740) | 0.830 (0.660–1.01) |
| Creatinine, mg/dL | 0.860 (0.590–1.08) | 0.690 (0.590–0.750) | 0.965 (0.800–1.08) |
| Blood urea nitrogen, mg/dL | 14.1 (9.70–23.0) | 12.2 (9.70–16.9) | 14.7 (10.5–23.0) |
| Alanine aminotransferase, U/L | 17.0 (3.00–74.0) | 10.5 (3.00–15.0) | 20.0 (10.0–74.0) |
| Aspartate aminotransferase, U/L | 21.0 (17.0–50.0) | 20.0 (17.0–21.0) | 24.0 (17.0–50.0) |
| Gamma-glutamyl transferase, U/L | 17.0 (9.00–69.0) | 12.5 (9.00–37.0) | 26.0 (15.0–69.0) |
| Renal functions | |||
| CrCl (mL/min) a | 105 (76.2–146) | 105 (85.0–122) | 106 (76.2–146) |
| BSA adjusted eGFRMDRD (mL/min) b | 93.2 (73.3–120) | 90.5 (81.5–111) | 95.4 (73.3–120) |
| BSA adjusted eGFRCKD-EPI_CR (mL/min) c | 108 (86.2–136) | 108 (98.9–115) | 110 (86.2–136) |
| BSA adjusted eGFRCKD-EPI_CC (mL/min) d | 110 (86.2–145) | 108 (104–112) | 113 (86.2–145) |
| BSA adjusted eGFRCKD-EPI_CRCC (mL/min) e | 111 (89.6–145) | 110 (104–118) | 116 (89.6–145) |
CrCl, creatinine clearance by Cockcroft-Gault equation; eGFR, estimated glomerular filtration rate; MDRD, modification of diet in renal disease; CKD-EPI, chronic kidney disease epidemiology collaboration; CR, creatinine; CC, cystatin C; min, the minimum of (CR or CC)/number and 1; max, the maximum of (CR or CC)/number and 1. a CrCl = (140 − Age) × weight/CR × 72 (×0.85 if female); b eGFR = 175 × CR−1.154 × Age−0.203 (×0.742 if female); c eGFR (female) = 142 × min (CR/0.7,1)−0.241 × max (CR/0.7,1)−1.200 × 0.9938Age × 1.012/1.73 × BSA; eGFR (male) = 142 × min (CR/0.9,1)−0.302 × max (CR/0.9,1)−1.200 × 0.9938Age /1.73 × BSA; d eGFR = 133 × min (CC/0.8,1)−0.499 × max (CC/0.8,1)−1.328 × 0.9962Age × 0.932 [if female]/1.73 × BSA; e eGFR (female) = 135 × min (CR/0.7,1)−0.219 × max (CR/0.7,1)−0.544 × min (CC/0.8,1)0.323 × max (CC/0.8,1)−0.778 × 0.9961Age × 0.963 /1.73 × BSA; eGFR (male) = 135 × min (CR/0.9,1)−0.144 × max (CR/0.9,1)−0.544 × min (CC/0.8,1)0.323 × max (CC/0.8,1)−0.778 × 0.9961Age/1.73 × BSA.
Table 2.
Parameter estimates and bootstrap medians (95% confidence intervals) for the final pharmacokinetic model of piperacillin in 12 healthy adults.
Table 2.
Parameter estimates and bootstrap medians (95% confidence intervals) for the final pharmacokinetic model of piperacillin in 12 healthy adults.
| Parameter | Estimates | RSE (%) [Shrinkage, %] | Bootstrap Median (95% CI) |
|---|---|---|---|
| Structural model | |||
| CL = θ1 × (CE /108.25) θ2 | |||
| θ1 (L/h) | 11.2 | 3.40 | 11.2 (10.5–12.1) |
| θ2 | 1.16 | 13.1 | 1.15 (0.811–1.59) |
| V1 = θ3 | |||
| θ3 (L) | 6.24 | 8.99 | 6.19 (5.27–7.57) |
| Q = θ4 × (LBM/50.08) θ5 | |||
| θ4 (L/h) | 4.32 a | ||
| θ5 | 2.50 | 13.9 | 2.45 (1.39–3.56) |
| V2 = θ6 × exp (θ7 × (WT − 61.7)) | |||
| θ6 (L) | 2.59 | 3.11 | 2.59 (2.28–2.74) |
| θ7 | 0.0288 | 8.38 | 0.0284 (0.0208–0.0371) |
| Interindividual variability | |||
| CL (%) | 7.17 | 30.3 [18.7] | 6.08 (0–10.5) |
| V1 (%) | 18.4 | 28.7 [19.1] | 17.5 (0–29.9) |
| Residual variability | |||
| Proportional error (%) | 13.4 | 12.2 [9.48] | 13.1 (9.39–16.0) |
RSE, relative standard error; CI, confidence interval; CL, total clearance; V1, central volume of distribution; V2, volume of distribution for the first peripheral compartment; Q, intercompartmental clearance between V1 and V2; CECR, estimated glomerular filtration rate by the CKD-EPI equation, utilizes serum creatinine levels; LBM, lean body mass; WT. weight; a, fixed.
Table 3.
Parameter estimates and bootstrap medians (95% confidence intervals) for the final pharmacokinetic model of tazobactam in 12 healthy adults.
Table 3.
Parameter estimates and bootstrap medians (95% confidence intervals) for the final pharmacokinetic model of tazobactam in 12 healthy adults.
| Parameter | Estimates | RSE (%) [Shrinkage, %] | Bootstrap Median (95% CI) |
|---|---|---|---|
| Structural model | |||
| CL = θ1 × (CE /108.25) θ2 | |||
| θ1 (L/h) | 12.4 | 3.26 | 12.3 (11.6–13.3) |
| θ2 | 0.857 | 13.1 | 0.858 (0.602–1.21) |
| V1 = θ3 | |||
| θ3 (L) | 9.03 | 6.40 | 9.02 (8.05–10.4) |
| Q = θ4 | |||
| θ4 (L/h) | 4.39 a | ||
| V2 = θ5 × exp (θ6 × (WT − 61.7)) | |||
| θ5 (L) | 3.21 | 5.48 | 3.23 (2.68–3.48) |
| θ6 | 0.0145 | 16.9 | 0.0142 (0.0106–0.0238) |
| Interindividual variability | |||
| CL (%) | 6.95 | 29.0 [7.37] | 6.17 (0.403–9.94) |
| Residual variability | |||
| Proportional error (%) | 13.5 | 9.57 [6.06] | 13.2 (10.4–15.6) |
RSE, relative standard error; CI, confidence interval; CL, total clearance; V1, central volume of distribution; V2, volume of distribution for the first peripheral compartment; Q, intercompartmental clearance between V1 and V2; CECR, estimated glomerular filtration rate by the CKD-EPI equation, utilizes serum creatinine levels; WT. weight; a, fixed.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Lee, Y.J.; Kang, G.; Zang, D.Y.; Lee, D.H. Population Pharmacokinetic Modeling of Piperacillin/Tazobactam in Healthy Adults and Exploration of Optimal Dosing Strategies. Pharmaceuticals 2025, 18, 1124. https://doi.org/10.3390/ph18081124
AMA Style
Lee YJ, Kang G, Zang DY, Lee DH. Population Pharmacokinetic Modeling of Piperacillin/Tazobactam in Healthy Adults and Exploration of Optimal Dosing Strategies. Pharmaceuticals. 2025; 18(8):1124. https://doi.org/10.3390/ph18081124
Chicago/Turabian StyleLee, Yun Jung, Gaeun Kang, Dae Young Zang, and Dong Hwan Lee. 2025. "Population Pharmacokinetic Modeling of Piperacillin/Tazobactam in Healthy Adults and Exploration of Optimal Dosing Strategies" Pharmaceuticals 18, no. 8: 1124. https://doi.org/10.3390/ph18081124
APA StyleLee, Y. J., Kang, G., Zang, D. Y., & Lee, D. H. (2025). Population Pharmacokinetic Modeling of Piperacillin/Tazobactam in Healthy Adults and Exploration of Optimal Dosing Strategies. Pharmaceuticals, 18(8), 1124. https://doi.org/10.3390/ph18081124
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
Article Metrics
Article metric data becomes available approximately 24 hours after publication online.
