Population pharmacokinetics of meropenem in critically ill infant patients

  • Wanlika Yonwises
    Affiliations
    Department of Pharmacy Practice, Faculty of Pharmaceutical Sciences, Chulalongkorn University, Bangkok, Thailand
    Search for articles by this author
  • Noppadol Wacharachaisurapol
    Affiliations
    Clinical Pharmacokinetics and Pharmacogenomics Research Unit, Department of Pharmacology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
    Search for articles by this author
  • Suvaporn Anugulruengkitt
    Affiliations
    Division of Infectious Diseases, Department of Pediatrics, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand

    Center of Excellence for Pediatric Infectious Diseases and Vaccines, Department of Pediatrics, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
    Search for articles by this author
  • Passara Maimongkol
    Affiliations
    Clinical Pharmacokinetics and Pharmacogenomics Research Unit, Department of Pharmacology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
    Search for articles by this author
  • Wanchai Treyaprasert
    Correspondence
    Corresponding author: Wanchai Treyaprasert, Ph.D., Department of Pharmacy Practice, Faculty of Pharmaceutical Sciences, Chulalongkorn University, 254 Phayathai Road, Pathumwan, Bangkok, Thailand, 10330. Telephone: +66 2218 8408
    Affiliations
    Department of Pharmacy Practice, Faculty of Pharmaceutical Sciences, Chulalongkorn University, Bangkok, Thailand
    Search for articles by this author
Open AccessPublished:August 18, 2021DOI:https://doi.org/10.1016/j.ijid.2021.08.031

      Abstract

      Background

      Population pharmacokinetic analysis in critically ill infants remains a challenge for lack of information.

      Objectives

      To determine the population pharmacokinetic parameters of meropenem and evaluate the covariates affecting population pharmacokinetic parameters.

      Methods

      A prospective study was conducted on 35 patients. A total of 160 blood samples were collected and determined free of drug concentrations of meropenem. Population pharmacokinetic data were analyzed using NONMEM software. Internal validation methods, including bootstrapping and prediction-corrected visual predictive checks, were applied to evaluate the robustness and predictive power of the final model.

      Results

      A one-compartment model with first-order elimination showed the best fit to the data.
      The typical clearance (CL) values and volume of distribution (Vd) were 1.33 L/h and 2.27 L, respectively. Weight and creatinine clearance were influential covariates for CL, while weight was a significant covariate for Vd of meropenem. The model evaluation results suggested robustness and good predictability of the final model. The standard dosage regimens of meropenem achieved 40% f T>MIC but not enough if a more aggressive target of 80% f T>MIC at MIC value of ≥ 16 µg/mL is desired.

      Conclusions

      This population pharmacokinetic model could be used for suggesting individualized meropenem dosage regimens in critically ill infants.

      Keywords

      Introduction

      The development of infections in critically ill patients is a dramatic problem since mortality and morbidity rates remain high. Antimicrobial therapy may not always be practical because pathophysiological changes associated with the course of the disease and treatment interventions may often alter drug pharmacokinetics (PK) (
      • Mattioli F
      • Fucile C
      • Del Bono V
      • Marini V
      • Parisini A
      • Molin A
      • et al.
      Population pharmacokinetics and probability of target attainment of meropenem in critically ill patients.
      ). Moreover, the intrinsic physicochemical properties of hydrophilic antimicrobials such as meropenem have to be considered at a much higher risk of inter-individual PK variations (
      • Pea F
      • Viale P
      • Furlanut M.
      Antimicrobial therapy in critically ill patients: a review of pathophysiological conditions responsible for altered disposition and pharmacokinetic variability.
      ).
      Meropenem is commonly administered for the treatment of severe infections in critically ill patients. Meropenem has a time-dependent bacterial killing characteristic, which means the free-drug concentration above the minimum inhibitory concentration (MIC) at least 40% of the time of the dosing interval (40% f T>MIC) is associated with optimal activity (
      • Braune S
      • König C
      • Roberts JA
      • Nierhaus A
      • Steinmetz O
      • Baehr M
      • et al.
      Pharmacokinetics of meropenem in septic patients on sustained low-efficiency dialysis: a population pharmacokinetic study.
      ). Understanding the PK of meropenem in the specific population is vital for establishing an effective dosage regimen. Previous studies also suggest that the PK of meropenem in critically ill patients differs from healthy volunteers (
      • Binder L
      • Schwörer H
      • Hoppe S
      • Streit F
      • Neumann S
      • Beckmann A
      • et al.
      Pharmacokinetics of meropenem in critically ill patients with severe infections.
      ,
      • Cies JJ
      • Moore WS
      • 2nd Enache A
      • Chopra A
      Population Pharmacokinetics and Pharmacodynamic Target Attainment of Meropenem in Critically Ill Young Children.
      ). Pathophysiological changes in critically ill patients have a profound effect on both volume of distribution (Vd) and clearance (CL) of meropenem (
      • Blot SI
      • Pea F
      • Lipman J.
      The effect of pathophysiology on pharmacokinetics in the critically ill patient—concepts appraised by the example of antimicrobial agents.
      ). Thus, developing a meropenem population PK model in critically ill patients may be considered a rational approach to optimize individual dosing regimens (
      • Crandon JL
      • Ariano RE
      • Zelenitsky SA
      • Nicasio AM
      • Kuti JL
      • Nicolau DP.
      Optimization of meropenem dosage in the critically ill population based on renal function.
      ,
      • Wang ZM
      • Chen XY
      • Bi J
      • Wang MY
      • Xu BP
      • Tang BH
      • et al.
      Reappraisal of the Optimal Dose of Meropenem in Critically Ill Infants and Children: a Developmental Pharmacokinetic-Pharmacodynamic Analysis.
      ). Data from the development population PK of meropenem in critically ill infants are limited (
      • Cies JJ
      • Moore WS
      • 2nd Enache A
      • Chopra A
      Population Pharmacokinetics and Pharmacodynamic Target Attainment of Meropenem in Critically Ill Young Children.
      ,
      • Kongthavonsakul K
      • Lucksiri A
      • Eakanunkul S
      • Roongjang S
      • Issaranggoon Na Ayuthaya S
      • Oberdorfer P
      Pharmacokinetics and pharmacodynamics of meropenem in children with severe infection.
      ,
      • Wang ZM
      • Chen XY
      • Bi J
      • Wang MY
      • Xu BP
      • Tang BH
      • et al.
      Reappraisal of the Optimal Dose of Meropenem in Critically Ill Infants and Children: a Developmental Pharmacokinetic-Pharmacodynamic Analysis.
      ). To our knowledge, a population PK analysis in critically ill infants has never been reported.
      Considering the high PK variability in critically ill patients and the lack of population PK information on meropenem in such patients, a population PK model of meropenem in Thai critically ill infants were performed in this study. This study aimed to determine the population PK parameters of meropenem and evaluate the covariates affecting population PK parameters in critically ill infants.

      Materials and Methods

       Patients and data collection

      A prospective, open-label, population pharmacokinetic study was conducted in the pediatric intensive care unit between February 2020 and February 2021 at King Chulalongkorn Memorial Hospital, a tertiary care teaching hospital in Thailand. This protocol was approved by the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University (IRB no. 565/62). It was given the Thai Clinical Trials Registry, registration number TCTR20191106001 (https://www.thaiclinicaltrials.org). Written informed consent was obtained from the parents of the infants before participating in the study.
      The inclusion criteria were aged one month to two years, receiving meropenem therapy and admitted to the pediatric intensive care unit (PICU); for patients admitted outside the PICU their criteria were as follows: (i) diagnosed with sepsis or septic shock (ii) had quick SOFA scores ≥ 2 (
      • Schlapbach LJ
      • Straney L
      • Bellomo R
      • MacLaren G
      • Pilcher D.
      Prognostic accuracy of age-adapted SOFA, SIRS, PELOD-2, and qSOFA for in-hospital mortality among children with suspected infection admitted to the intensive care unit.
      ) (iii) receiving the vasopressor drug, and (iv) using invasive mechanical ventilation. Patients were excluded from the study if they met one of the following criteria: (i) known or suspected hypersensitivity to meropenem (ii) weight < 3,000 g (iii) preterm birth history (iv) estimated creatinine clearance (CLCR) < 50 mL/min/1.73 m2 (v) receiving renal replacement therapy or kidney transplant (vi) receiving extracorporeal membrane oxygenation and (vii) receiving co-administration of valproic acid, probenecid, and ceftriaxone.
      Data collection from critically ill infants included patient demographic characteristics (gender, age, weight, and height), diagnosis, comorbidities, the dosage of meropenem, concomitant medications (e.g., inotropic drug and diuretic drug), biochemistry (e.g., serum creatinine), presence of invasive mechanical ventilation and other factors.

       Dosing regimen and blood sampling

      Meropenem (MapenemTM: Siam Pharmaceutical Co., Ltd., Bangkok, Thailand) was administered as an intravenous infusion over 1 or 3 h at 20-40 mg/kg/dose every 8 h. Blood samples were collected at a steady-state after administration of the second dose of meropenem.
      Blood samples were drawn from a central venous catheter at 1 (1-h infusion regimen only), 2, 3 (3-h infusion regimen only), 4, 6, and 8 h post-dose, resulting in a total of five blood samples per patient. All blood samples were collected into heparinized tubes and centrifuged at 4,000 rpm for ten minutes. The serum was separated and stored at −80 °C until analysis.

       Meropenem concentration determination

      The free drug concentrations of meropenem were quantified using a validated high-performance liquid chromatography with ultraviolet (HPLC-UV) detector (Shimadzu, Kyoto, Japan) (
      • Briscoe SE
      • McWhinney BC
      • Lipman J
      • Roberts JA
      • Ungerer JP.
      A method for determining the free (unbound) concentration of ten beta-lactam antibiotics in human plasma using high performance liquid chromatography with ultraviolet detection.
      ) at the Department of Pharmacology, Faculty of Medicine, Chulalongkorn University. The chromatographic separation was performed with a Luna 5 µm C18, LC Column 250 × 4.6 mm (Phenomenex, California, United States of America). The UV detection of standard and internal standard were performed at 304 nm and 250 nm, respectively; 400 µL of the plasma samples was subjected to ultrafiltration using an Amicon Ultra-0.5 mL centrifugal filter device with 30,000 molecular weight cut-offs (Millipore, Cork, Ireland). The device was centrifuged for 30 min at 15,600 × g. An aliquot (200 L) of the ultrafiltrate was then transferred to an autosampler vial and vortex mixed for 30 s with 10 µL of ampicillin sodium and 10 µL of 1.0 M MES buffer (pH 6.6) before proceeding with the chromatographic analysis. The standard curve for the meropenem bioassay ranged from 0.5 to 50 µg/mL, and the regression coefficients (r2) were 0.999. The assays' accuracy and precision were evaluated by calculating the lower limit of quantification (LLOQ) using quality control samples at low, medium, and high concentrations of the calibration ranges. All results met the bioanalysis acceptance criteria of the US Food and Drugs Administration.

       Population PK analysis

      The population PK of meropenem was analyzed by non-linear mixed-effect modeling using NONMEM version 7.4.3 (Icon Development Solutions, Ellicott City, MD, USA). The NONMEM runs were executed with PDx-Pop version 5.2.1 (Icon Development Solutions, Ellicott City, MD, USA). The usual first-order conditional estimation with interaction (FOCE-I) method was used throughout. One- and two-compartment models with first-order elimination were evaluated as the base structural model. The interindividual variability in the PK parameters was evaluated using additive, proportional, and exponential error models. Additive, proportional, exponential, and combined proportional and additive error models were used to access the residual variability.
      Model discrimination was performed by the objective function value (OFV) and the Akaike information criterion (AIC).
      Covariate analysis was performed after the selection of the base model. For covariate screening, potential covariates were selected based on physiological plausibility and prior knowledge. The influence of each covariate was first screened using a scatterplot of a parameter versus covariates. Subsequently, direct covariate testing was conducted using the stepwise method to establish the full and final model.
      The covariates considered for testing included sex, age, body weight (WT), CLCR estimated with the Bedside Schwartz equation (
      • Schwartz GJ
      • Work DF.
      Measurement and estimation of GFR in children and adolescents.
      ), use of mechanical ventilation, use of vasoactive medications, sepsis or septic shock, and congestive heart failure. The continuous covariates (such as age, WT, and CLCR) were centered on their median values and were tested via linear, power, and exponential models. The categorical covariates (such as sex, use of mechanical ventilation, use of vasoactive medications, sepsis or septic shock, and congestive heart failure) were also examined with linear, proportional, power, and exponential models. During the covariate model-building process, stepwise forward inclusion and backward elimination approaches were employed. Reductions in the OFV of at least 3.84 (p < 0.05) and of greater than 6.64 (p < 0.01) were required for a covariate to be considered significant in the forward inclusion and backward elimination steps, respectively.

       Model Evaluation

      Visual evaluation methods (goodness-of-fit plots) were applied to evaluate the performance of both the base and final models. The precision of the parameter estimates was expressed as relative standard error (RSE, %) and confidence intervals (CI). The RSE were directly computed by NONMEM, and a value < 30% for fixed effects and < 50% for random effects were considered acceptable (
      • Alvarez JC
      • Moine P
      • Davido B
      • Etting I
      • Annane D
      • Larabi IA
      • et al.
      Population pharmacokinetics of lopinavir/ritonavir in Covid-19 patients.
      ). A nonparametric bootstrap method was used to verify the robustness of standard approximations for parameter uncertainty of the final model. One thousand bootstrap data sets were generated by resampling from the original data set. Median parameter values and the 2.5th-97.5th percentile from bootstrap estimates were compared with the final model estimates. A prediction-corrected visual predictive check (pcVPC) was also performed by simulating 1,000 patients to evaluate the predictive performance of the final model. Visual checks were performed by overlaying the observed data points with the 95%CI of the simulated 5th, 50th, and 95th percentile curves.

      Results

       Patient characteristics and clinical data

      A total of 35 critically ill infant patients (16 males and 19 females) were included in the study. Of these patients, the median age was 7.53 months (range 1.13-22.50 months), median WT was 5.70 kg (range 3.00-15.02 kg), and the median CLCR was 112.43 mL/min/1.73 m2 (range 63.54-239.20 mL/min/1.73 m2). Seventeen patients (48.57%) received vasoactive medications, and 19 patients (54.29%) received mechanical ventilation. The demographics of patients are shown in Table 1.
      Table 1Demographic characteristics of the study patients.
      CharacteristicsValue
      Data are expressed as median (interquartile range) or n(%).
      Female (%)

      Age (months)

      Weight (kg)

      Creatinine clearance (mL/min/1.73 m2)

      Sepsis or septic shock (%)

      Congestive heart failure (%)

      The use of vasoactive medications (%)

      The use mechanical ventilation (%)
      19 (54.29)

      7.53 (1.13-22.50)

      5.70 (3.00-15.02)

      112.43 (63.54-239.20)

      16 (45.71)

      4 (11.76)

      17 (48.57)

      19 (54.29)
      a Data are expressed as median (interquartile range) or n(%).

       Population PK model

      The free meropenem concentration-time profile is shown in Figure 1. A one-compartment model best described the data with first-order elimination as demonstrated by an OFV of 776.767 and AIC of 788.767 versus an OFV of 815.672 and AIC of 835.672 for the comparison between a one-compartment and two-compartment model, respectively. Interindividual variability was modeled using an exponential model. Residual variability was described with a combined proportional and additive error model.
      Figure 1
      Figure 1Meropenem concentrations (unbound) of critically ill infant patients.
      Gender (SEX), age, WT, CLCR, sepsis or septic shock (SEPSK), congestive heart failure (CHF), vasoactive medications (VASD), and mechanical ventilation (VEN) were selected as candidates for the covariate to use in the covariate model. The examined covariates, SEX, SEPSK, CHF, and VASD, did not show significant covariate effects (p > 0.05) on any of the PK parameters. When age and WT were incorporated into the basic model for each PK parameter, all of them markedly improved the fit (decrease of OFV greater than 3.84, p < 0.05). Because of the high collinearity (r = 0.733) between age and WT, only WT that produced the most significant decrease of OFV was added on each of the PK parameters to avoid a collinearity effect (
      • Du X
      • Li C
      • Kuti JL
      • Nightingale CH
      • Nicolau DP.
      Population pharmacokinetics and pharmacodynamics of meropenem in pediatric patients.
      ).
      The resultant full model included CLCR, WT, and VEN as covariates for CL, WT for Vd. During the backward elimination process to find the final model, it was found that VEN could be excluded from the full model without causing a significant increase of the OFV (4.313, p > 0.01). The model building process is summarized in Table 2.
      Table 2Summary of the covariate models development.
      ProcessOFVΔ OFV
      Δ OFV represents changed value of OFV in the model compared with the value in the model with addition or elimination of covariate.
      Stepwise forward addition

       Structure model

      WT impact on CL

      WT impact on Vd

      CLCR impact on CL

      VEN impact on CL

      Stepwise backward elimination

       Full model

      WT impact on CL

      WT impact on Vd

      CLCR impact on CL




      731.677

      755.595

      713.392

      709.079





      759.190

      740.560

      731.677




      -23.918

      -21.172

      -18.285

      -4.313





      +45.798

      +27.168

      +18.285
      a Δ OFV represents changed value of OFV in the model compared with the value in the model with addition or elimination of covariate.
      The population parameter estimates of the final model for CL and Vd were 1.33 L/h and 2.27 L, respectively. The effect of covariates on meropenem CL and Vd were included as shown in equations (1) and (2).
      CL = 1.33 x exp(0.126 x (WT- 5.70)) + 0.0085 x (CLCR -112.43) (1)
      Vd = 2.27 x exp(0.139 x (WT-5.70)) (2)
      Where CL is clearance (L/h), Vd is the value of the volume of distribution (L). In the formula, WT and CLCR represent body weight and creatinine clearance, respectively.
      After incorporating significant covariates in the final model, interindividual variability concerning the basic model was reduced from 59.2% to 28.95% as CV for CL and from 48.6% to 28.02% as CV for Vd, respectively.

       Model evaluation

      Model diagnostics showed acceptable goodness-of-fit for the final model of meropenem.
      It was observed that the scatterplots of PRED and IPRED versus observed concentration symmetric, distributed around the identity line (Figure 2A and B). Besides, the scatterplot of the population predicted concentration (PRED) and time after dose versus conditional weighted residuals (CWRES), demonstrated a good distribution of the point around the zero lines. Most of the points were within the range of -2 and 2, indicating that the model was significantly well fitted (Figure 2C and D). Bootstrap analysis was performed to assess the accuracy and robustness of the estimation parameters from the final model. The 1,000 bootstrap data set was generated by resampling with a success rate of 99.9%. The result of the bootstrap analysis is shown in Table 3. The median values of parameter estimates from the bootstrap were close to the population estimates in the final model. The significance of covariates was further verified by the results showing that all parameters' 95% CI did not include null. The symmetric 95% CI of the final model was also consistent with the 2.5th-97.5th percentile of bootstrap estimates, indicating the accuracy and robustness of the proposed model. The result of the pcVPC is shown in Figure 3. As seen in
      Figure 2
      Figure 2The goodness-of-fit plots of the final model. (A) observed versus population predicted concentrations; (B) observed versus individual predicted concentrations; (C) conditional weighted residuals (CWRES) versus population predicted concentrations; (D) conditional weighted residuals (CWRES) versus time.
      Table 3Population PK parameter estimates of base model, final model, and bootstrap analysis.
      ParametersBase model Estimate (RSE, %), 95% CIFinal model Estimate (RSE, %), 95% CIBootstrap Median (Range
      (2.5th-97.5th percentiles) of 1,000 bootstrap.
      )
      CL (L/hr)



      Vd (L)



      θWT, CL



      θCLCR, CL



      θWT, Vd



      Interindividual variability

      CL (CV, %)



      Vd (CV, %)

      Residual variability

      proportional (CV, %)



      additive (µg/mL)
      1.47 (10.7),

      1.16-1.78

      2.58 (9.61),

      2.09-3.07

      -



      -



      -





      59.2, (24.0),

      43.13-71.83

      48.6, (31.9),

      29.77-61.89



      27.13, (23.4),

      20.07-32.71

      0.474, (38.3),

      0.237-0.628
      1.33 (6.69),

      1.16-1.50

      2.27 (7.93),

      1.92-2.62

      0.126 (14.7),

      0.0850-0.167

      0.00850 (16.6),

      0.00458-0.0124

      0.139 (23.5),

      0.0990-0.179



      28.95, (36.4),

      15.49-37.95

      28.02, (41.0),

      12.41-37.68



      28.0, (23.1),

      20.71-33.76

      0.463, (35.0),

      0.259-0.601
      1.33

      (1.16-1.52)

      2.28

      (1.93-2.67)

      0.121

      (0.0577-0.162)

      0.00867

      (0.00442-0.0132)

      0.141

      (0.0999-0.196)



      27.62

      (16.46-38.47)

      31.30

      (12.00-37.82)



      27.97

      (20.88-34.64)

      0.457

      (0.0985-0.834)
      CL, clearance; Vd, volume of distribution; WT, weight;
      CLCR, creatinine clearance (mL/min/1.73 m2); CV, coefficient of variation;
      RSE, relative standard error; CI: confident interval
      a (2.5th-97.5th percentiles) of 1,000 bootstrap.
      Figure 3
      Figure 3Prediction-corrected visual predictive check (n = 1,000 simulations) of the final model. The solid lines represent the 5th, 50th, and 95th percentiles of the observed concentrations; the 3 shaded areas represent the 90% confidence intervals for corresponding percentiles.
      Figure 3, less than 10% (5.23%) of observed data points fall outside a 90% prediction interval.
      It was demonstrated that the model and parameter estimates adequately described the observed data. Furthermore, the pcVPC with the final covariate model confirmed the model's goodness-of-fit to the observed data.
      The calculated results of %fT>MIC for meropenem in critically ill infants were conducted. The prediction of %fT>MIC for various meropenem regimens against the MICs of 1 µg/mL, 2 µg/mL, 4 µg/mL, 8 µg/mL, and 16 µg/mL are shown in Table 4.
      Table 4Dosage regimens and % f T>MIC for meropenem in critically ill patients in this study.
      Weight- based dosing

      q 8 h (mg per dose)
      %f T>MIC
      MIC 1 (µg/mL)MIC 2 (µg/mL)MIC 4 (µg/mL)MIC 8 (µg/mL)MIC 16 (µg/mL)
      Infuse (h)Infuse (h)Infuse (h)Infuse (h)Infuse (h)
      1313131313
      60

      90

      100

      120

      150

      160

      200

      240

      300

      320

      360

      400

      450

      480

      600
      77

      85

      88

      91

      96

      98

      100

      100

      100

      100

      100

      100

      100

      100

      100
      86

      94

      96

      100

      100

      100

      100

      100

      100

      100

      100

      100

      100

      100

      100
      61

      70

      73

      77

      81

      83

      88

      91

      96

      98

      100

      100

      100

      100

      100
      69

      79

      82

      86

      90

      92

      96

      100

      100

      100

      100

      100

      100

      100

      100
      46

      55

      57

      61

      67

      68

      73

      77

      81

      83

      85

      88

      90

      91

      96
      50

      61

      65

      69

      74

      75

      82

      86

      90

      92

      94

      96

      99

      100

      100
      28

      38

      41

      46

      51

      52

      57

      61

      67

      68

      70

      73

      75

      77

      81
      21

      40

      45

      50

      57

      58

      65

      69

      74

      75

      79

      82

      84

      86

      90
      9

      21

      24

      28

      35

      36

      41

      46

      51

      52

      55

      57

      60

      61

      67
      0

      0

      0

      21

      33

      36

      45

      50

      57

      58

      61

      65

      67

      69

      74

      Discussion

      In the present study, a one-compartment PK model with first-order elimination was appropriate to describe PK data, consistent with previous studies (
      • Bradley JS
      • Sauberan JB
      • Ambrose PG
      • Bhavnani SM
      • Rasmussen MR
      • Capparelli EV.
      Meropenem pharmacokinetics, pharmacodynamics, and Monte Carlo simulation in the neonate.
      ,
      • Germovsek E
      • Lutsar I
      • Kipper K
      • Karlsson MO
      • Planche T
      • Chazallon C
      • et al.
      Plasma and CSF pharmacokinetics of meropenem in neonates and young infants: results from the NeoMero studies.
      ,
      • Smith PB
      • Cohen-Wolkowiez M
      • Castro LM
      • Poindexter B
      • Bidegain M
      • Weitkamp JH
      • et al.
      Population pharmacokinetics of meropenem in plasma and cerebrospinal fluid of infants with suspected or complicated intra-abdominal infections.
      ,
      • van den Anker JN
      • Pokorna P
      • Kinzig-Schippers M
      • Martinkova J
      • de Groot R
      • Drusano GL
      • et al.
      Meropenem pharmacokinetics in the newborn.
      ). In the final model, CL of meropenem was influenced by CLCR and WT. The most significant covariate of meropenem apparent CL was CLCR. This finding is consistent with previous studies (
      • Du X
      • Li C
      • Kuti JL
      • Nightingale CH
      • Nicolau DP.
      Population pharmacokinetics and pharmacodynamics of meropenem in pediatric patients.
      ,
      • Parker EM
      • Hutchison M
      • Blumer JL.
      The pharmacokinetics of meropenem in infants and children: a population analysis.
      ,
      • Wang ZM
      • Chen XY
      • Bi J
      • Wang MY
      • Xu BP
      • Tang BH
      • et al.
      Reappraisal of the Optimal Dose of Meropenem in Critically Ill Infants and Children: a Developmental Pharmacokinetic-Pharmacodynamic Analysis.
      ). The explanation is that meropenem is a hydrophilic drug and is eliminated by glomerular filtration (
      • Nicolau DP.
      Pharmacokinetic and Pharmacodynamic Properties of Meropenem.
      ). The effect of altered kidney function on meropenem plasma concentration is demonstrated from the simulation (Figure 4B), as shown in the discussion. At the same meropenem dose administered, patients with impaired kidney function had a higher meropenem level, while patients with augmented kidney clearance had a lower meropenem level. WT was also a significant covariate for meropenem CL and could be explained by the correlation between WT and the age of patients. In this study, patients with greater WT indicated older age and were more likely to have a mature kidney function comparing patients with lesser WT.
      Figure 4
      Figure 4Simulated concentration profiles of meropenem for critically ill infant patients receiving 40 mg/kg q 8 h (A) patients with normal renal function administered as intravenous 1 h and 3 h infusion (B) patients with abnormal renal function administered as an intravenous 3 h infusion.
      In agreement with several previous studies (
      • Du X
      • Li C
      • Kuti JL
      • Nightingale CH
      • Nicolau DP.
      Population pharmacokinetics and pharmacodynamics of meropenem in pediatric patients.
      ,
      • Ikawa K
      • Morikawa N
      • Ikeda K
      • Miki M
      • Kobayashi M.
      Population pharmacokinetics and pharmacodynamics of meropenem in Japanese pediatric patients.
      ,
      • Ohata Y
      • Tomita Y
      • Nakayama M
      • Kozuki T
      • Sunakawa K
      • Tanigawara Y.
      Optimal dosage regimen of meropenem for pediatric patients based on pharmacokinetic/pharmacodynamic considerations.
      ,
      • Parker EM
      • Hutchison M
      • Blumer JL.
      The pharmacokinetics of meropenem in infants and children: a population analysis.
      ,
      • Wang ZM
      • Chen XY
      • Bi J
      • Wang MY
      • Xu BP
      • Tang BH
      • et al.
      Reappraisal of the Optimal Dose of Meropenem in Critically Ill Infants and Children: a Developmental Pharmacokinetic-Pharmacodynamic Analysis.
      ), we found that there was a relationship between Vd of meropenem and WT. The mean of the estimated CL and Vd values were 1.33 L/h and 2.27 L, respectively. The median estimated weight-normalized CL and Vd were 0.23 L/h/kg and 0.4 L/kg, respectively. They were similar to the values reported previously (
      • Blumer JL
      • Reed MD
      • Kearns GL
      • Jacobs RF
      • Gooch 3rd, WM
      • Yogev R
      • et al.
      Sequential, single-dose pharmacokinetic evaluation of meropenem in hospitalized infants and children.
      ,
      • Parker EM
      • Hutchison M
      • Blumer JL.
      The pharmacokinetics of meropenem in infants and children: a population analysis.
      ,
      • Smith PB
      • Cohen-Wolkowiez M
      • Castro LM
      • Poindexter B
      • Bidegain M
      • Weitkamp JH
      • et al.
      Population pharmacokinetics of meropenem in plasma and cerebrospinal fluid of infants with suspected or complicated intra-abdominal infections.
      ). The estimated CL in this study was lower than in several other studies (
      • Cies JJ
      • Moore WS
      • 2nd Enache A
      • Chopra A
      Population Pharmacokinetics and Pharmacodynamic Target Attainment of Meropenem in Critically Ill Young Children.
      ,
      • Kongthavonsakul K
      • Lucksiri A
      • Eakanunkul S
      • Roongjang S
      • Issaranggoon Na Ayuthaya S
      • Oberdorfer P
      Pharmacokinetics and pharmacodynamics of meropenem in children with severe infection.
      ,
      • Wang ZM
      • Chen XY
      • Bi J
      • Wang MY
      • Xu BP
      • Tang BH
      • et al.
      Reappraisal of the Optimal Dose of Meropenem in Critically Ill Infants and Children: a Developmental Pharmacokinetic-Pharmacodynamic Analysis.
      ) when stratified according to age and WT. Kongthavonsakul and colleagues (
      • Kongthavonsakul K
      • Lucksiri A
      • Eakanunkul S
      • Roongjang S
      • Issaranggoon Na Ayuthaya S
      • Oberdorfer P
      Pharmacokinetics and pharmacodynamics of meropenem in children with severe infection.
      ) reported that the mean value of CL was 6.53 L/h in 14 children (aged 4.5 to 11.8 years). Cies et al. (
      • Cies JJ
      • Moore WS
      • 2nd Enache A
      • Chopra A
      Population Pharmacokinetics and Pharmacodynamic Target Attainment of Meropenem in Critically Ill Young Children.
      ) suggested the median CL was 0.42 L/h/kg in nine critically ill young children (aged one to nine years). Wang et al. (
      • Wang ZM
      • Chen XY
      • Bi J
      • Wang MY
      • Xu BP
      • Tang BH
      • et al.
      Reappraisal of the Optimal Dose of Meropenem in Critically Ill Infants and Children: a Developmental Pharmacokinetic-Pharmacodynamic Analysis.
      ) reported that the median CL was 0.43 L/h/kg in 57 critically ill infants and children (aged 0.101 to 14.4 years). A trend of increased CL was observed with the increasing age of the patients due to impaired kidney function during the first two years of life due to normal growth and development (
      • Rodieux F
      • Wilbaux M
      • van den Anker JN
      • Pfister M.
      Effect of Kidney Function on Drug Kinetics and Dosing in Neonates, Infants, and Children.
      ). Our study investigated data in critically ill infant patients with congenital heart disease 47.06% and congestive heart failure 11.76%. Congenital heart disease can be associated with renal injury and dysfunction (
      • Zheng J
      • Yao Y
      • Han L
      • Xiao Y.
      Renal function and injury in infants and young children with congenital heart disease.
      ).
      The estimated Vd of 0.4 L/kg in this investigation seems comparable to the values reported in previous studies (
      • Blumer JL
      • Reed MD
      • Kearns GL
      • Jacobs RF
      • Gooch 3rd, WM
      • Yogev R
      • et al.
      Sequential, single-dose pharmacokinetic evaluation of meropenem in hospitalized infants and children.
      ,
      • Bradley JS
      • Sauberan JB
      • Ambrose PG
      • Bhavnani SM
      • Rasmussen MR
      • Capparelli EV.
      Meropenem pharmacokinetics, pharmacodynamics, and Monte Carlo simulation in the neonate.
      ,
      • Smith PB
      • Cohen-Wolkowiez M
      • Castro LM
      • Poindexter B
      • Bidegain M
      • Weitkamp JH
      • et al.
      Population pharmacokinetics of meropenem in plasma and cerebrospinal fluid of infants with suspected or complicated intra-abdominal infections.
      ). The estimated Vd differs from the value of Vd in the previous populations, including neonates, young infants, infants, and children (
      • Germovsek E
      • Lutsar I
      • Kipper K
      • Karlsson MO
      • Planche T
      • Chazallon C
      • et al.
      Plasma and CSF pharmacokinetics of meropenem in neonates and young infants: results from the NeoMero studies.
      ,
      • Ikawa K
      • Morikawa N
      • Ikeda K
      • Miki M
      • Kobayashi M.
      Population pharmacokinetics and pharmacodynamics of meropenem in Japanese pediatric patients.
      ,
      • Kongthavonsakul K
      • Lucksiri A
      • Eakanunkul S
      • Roongjang S
      • Issaranggoon Na Ayuthaya S
      • Oberdorfer P
      Pharmacokinetics and pharmacodynamics of meropenem in children with severe infection.
      ,
      • Ohata Y
      • Tomita Y
      • Nakayama M
      • Kozuki T
      • Sunakawa K
      • Tanigawara Y.
      Optimal dosage regimen of meropenem for pediatric patients based on pharmacokinetic/pharmacodynamic considerations.
      ). The data from the Ikawa et al. (
      • Ikawa K
      • Morikawa N
      • Ikeda K
      • Miki M
      • Kobayashi M.
      Population pharmacokinetics and pharmacodynamics of meropenem in Japanese pediatric patients.
      ) and Ohata et al. (
      • Ohata Y
      • Tomita Y
      • Nakayama M
      • Kozuki T
      • Sunakawa K
      • Tanigawara Y.
      Optimal dosage regimen of meropenem for pediatric patients based on pharmacokinetic/pharmacodynamic considerations.
      ) studies were collected from infants and children and suggest the Vd are 0.012 L/kg and 0.019 L/kg, respectively. Germovsek et al. (
      • Germovsek E
      • Lutsar I
      • Kipper K
      • Karlsson MO
      • Planche T
      • Chazallon C
      • et al.
      Plasma and CSF pharmacokinetics of meropenem in neonates and young infants: results from the NeoMero studies.
      ) suggested the value of Vd was 0.259 L/kg in neonates and young infants. Kongthavonsakul et al. (
      • Kongthavonsakul K
      • Lucksiri A
      • Eakanunkul S
      • Roongjang S
      • Issaranggoon Na Ayuthaya S
      • Oberdorfer P
      Pharmacokinetics and pharmacodynamics of meropenem in children with severe infection.
      ) investigated children with a severe infection and reported the value of Vd was 0.097 L/kg. The Vd of drugs changes in children as they age. These age-related changes are due to changes in the total body water spaces such that extracellular water decreases during development, from 70% total body weight in newborns to 61.2% in 1-year-old infants (
      • Batchelor HK
      • Marriott JF.
      Paediatric pharmacokinetics: key considerations.
      ). Our data suggest a larger Vd estimate for meropenem than previously mentioned because this study collected data in critically ill patients with sepsis and septic shock who sustain capillary leakage out of the blood and who receive intravenous fluids (
      • Blot SI
      • Pea F
      • Lipman J.
      The effect of pathophysiology on pharmacokinetics in the critically ill patient—concepts appraised by the example of antimicrobial agents.
      ).
      The population PK model developed in this study has helped establish dosage regimens of meropenem administered over 1-h and 3-h infusions. Based on the current population analysis results, we simulated the concentration-time curves for critically ill infants (Figure 4). The results suggest that the time of free drug concentrations above the MIC should be prolonged by 3-h infusion administration, as shown in Figure 4A. Figure 4B shows the effects of CLCR on the typical population PK profiles of meropenem for each type of patient and under the same dosing schedule used in the current study. It is found that CLCR has a much more significant impact on plasma disposition of meropenem in patients with renal insufficiency (CLCR < 60 mL/min/1.73 m2) and augmented renal clearance (CLCR > 130 mL/min/1.73 m2). The results of % f T>MIC at various MIC values, calculated as various regimens, are summarized in Table 4. Meropenem dosage regimens as weight-based dosing included 20, 30, and 40 mg/kg every 8 h as a 1-h and 3-h infusion in patients with the weight of 3, 5, 8, 10, 12, and 15 kg, as a weight range of 3 to 15 kg was seen in this population. It was found that a dose of 60 mg appears adequate to achieve 40% f T>MIC at MIC values of 1 µg/mL, 2 µg/mL, and 4 µg/mL. If the MICs were to increase to 8 and 16 µg/mL, it would be necessary to increase the dose to 100 mg and 200 mg to achieve the target %f T>MIC value of > 40, respectively. For a more aggressive target of 80% f T>MIC, a dose of 200 mg would not be enough to treat patients with MIC of 16 µg/mL. Yu et al. (
      • Yu Z
      • Pang X
      • Wu X
      • Shan C
      • Jiang S.
      Clinical outcomes of prolonged infusion (extended infusion or continuous infusion) versus intermittent bolus of meropenem in severe infection: A meta-analysis.
      ) suggest that the extended infusion can prolong the % f T>MIC and improve antibacterial activity. 24-h continuous infusion regimens may be necessary to achieve 80% f T>MIC. Based on these findings, we believe that a dose of 60 mg can provide appropriate pharmacodynamic exposure of 40% f T>MIC for Gram-negative organisms up to breakpoint for Pseudomonas aeruginosa of 4 µg/mL. However, confirmation with a clinical investigation in critically ill infants is warranted.
      Our study is, to our knowledge, the first report on the population PK of meropenem in critically ill infants aged one month to two years. Population PK and pharmacodynamic exposure are directly analyzed from free drug concentrations. This study has some limitations. The final model of meropenem was only internally validated; external validation was not implemented due to the limited number of patients. In addition, the efficacy and safety of simulated meropenem dose regimens were not evaluated in the current study.

      Conclusions

      A population PK model of meropenem in critically ill infants was developed and validated. It was found that the clearance of meropenem was correlated with creatinine clearance and body weight, whereas the volume of distribution was correlated with body weight. This population PK model could be used for suggesting individualized meropenem dosage regimens in critically ill infants.

      Funding

      This research is funded by Chulalongkorn University (grant number CU_GR_63_15-_33_04).

      Ethical approval

      This protocol was approved by the Institutional Review Board of Faculty of Medicine, Chulalongkorn University (IRB no. 565/62).

      Declaration of interests

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Acknowledgments

      The authors would like to thank Thanakit Pongpitakmetha, MD and Surachai Leksuwankun, MD, and the PICU team (physicians and nurses), for their support. We also thank the Clinical Research Laboratory of the Faculty of Medicine, Chulalongkorn University, and the Clinical Pharmacokinetics and Pharmacogenomics Research Unit of the Department of Pharmacology, Faculty of Medicine, Chulalongkorn University for blood sample preparation and analysis.

      References

        • Alvarez JC
        • Moine P
        • Davido B
        • Etting I
        • Annane D
        • Larabi IA
        • et al.
        Population pharmacokinetics of lopinavir/ritonavir in Covid-19 patients.
        Eur J Clin Pharmacol. 2021; 77: 389-397
        • Batchelor HK
        • Marriott JF.
        Paediatric pharmacokinetics: key considerations.
        Br J Clin Pharmacol. 2015; 79: 395-404
        • Binder L
        • Schwörer H
        • Hoppe S
        • Streit F
        • Neumann S
        • Beckmann A
        • et al.
        Pharmacokinetics of meropenem in critically ill patients with severe infections.
        Ther Drug Monit. 2013; 35: 63-70
        • Blot SI
        • Pea F
        • Lipman J.
        The effect of pathophysiology on pharmacokinetics in the critically ill patient—concepts appraised by the example of antimicrobial agents.
        Advanced drug delivery reviews. 2014; 77: 3-11
        • Blumer JL
        • Reed MD
        • Kearns GL
        • Jacobs RF
        • Gooch 3rd, WM
        • Yogev R
        • et al.
        Sequential, single-dose pharmacokinetic evaluation of meropenem in hospitalized infants and children.
        Antimicrob Agents Chemother. 1995; 39: 1721-1725
        • Bradley JS
        • Sauberan JB
        • Ambrose PG
        • Bhavnani SM
        • Rasmussen MR
        • Capparelli EV.
        Meropenem pharmacokinetics, pharmacodynamics, and Monte Carlo simulation in the neonate.
        Pediatr Infect Dis J. 2008; 27: 794-799
        • Braune S
        • König C
        • Roberts JA
        • Nierhaus A
        • Steinmetz O
        • Baehr M
        • et al.
        Pharmacokinetics of meropenem in septic patients on sustained low-efficiency dialysis: a population pharmacokinetic study.
        Crit Care. 2018; 22: 25
        • Briscoe SE
        • McWhinney BC
        • Lipman J
        • Roberts JA
        • Ungerer JP.
        A method for determining the free (unbound) concentration of ten beta-lactam antibiotics in human plasma using high performance liquid chromatography with ultraviolet detection.
        J Chromatogr B Analyt Technol Biomed Life Sci. 2012; 907: 178-184
        • Cies JJ
        • Moore WS
        • 2nd Enache A
        • Chopra A
        Population Pharmacokinetics and Pharmacodynamic Target Attainment of Meropenem in Critically Ill Young Children.
        J Pediatr Pharmacol Ther. 2017; 22: 276-285
        • Crandon JL
        • Ariano RE
        • Zelenitsky SA
        • Nicasio AM
        • Kuti JL
        • Nicolau DP.
        Optimization of meropenem dosage in the critically ill population based on renal function.
        Intensive care medicine. 2011; 37: 632-638
        • Du X
        • Li C
        • Kuti JL
        • Nightingale CH
        • Nicolau DP.
        Population pharmacokinetics and pharmacodynamics of meropenem in pediatric patients.
        J Clin Pharmacol. 2006; 46: 69-75
        • Germovsek E
        • Lutsar I
        • Kipper K
        • Karlsson MO
        • Planche T
        • Chazallon C
        • et al.
        Plasma and CSF pharmacokinetics of meropenem in neonates and young infants: results from the NeoMero studies.
        J Antimicrob Chemother. 2018; 73: 1908-1916
        • Ikawa K
        • Morikawa N
        • Ikeda K
        • Miki M
        • Kobayashi M.
        Population pharmacokinetics and pharmacodynamics of meropenem in Japanese pediatric patients.
        J Infect Chemother. 2010; 16: 139-143
        • Kongthavonsakul K
        • Lucksiri A
        • Eakanunkul S
        • Roongjang S
        • Issaranggoon Na Ayuthaya S
        • Oberdorfer P
        Pharmacokinetics and pharmacodynamics of meropenem in children with severe infection.
        Int J Antimicrob Agents. 2016; 48: 151-157
        • Mattioli F
        • Fucile C
        • Del Bono V
        • Marini V
        • Parisini A
        • Molin A
        • et al.
        Population pharmacokinetics and probability of target attainment of meropenem in critically ill patients.
        Eur J Clin Pharmacol. 2016; 72: 839-848
        • Nicolau DP.
        Pharmacokinetic and Pharmacodynamic Properties of Meropenem.
        Clinical Infectious Diseases. 2008; 47: S32-S40
        • Ohata Y
        • Tomita Y
        • Nakayama M
        • Kozuki T
        • Sunakawa K
        • Tanigawara Y.
        Optimal dosage regimen of meropenem for pediatric patients based on pharmacokinetic/pharmacodynamic considerations.
        Drug Metab Pharmacokinet. 2011; 26: 523-531
        • Parker EM
        • Hutchison M
        • Blumer JL.
        The pharmacokinetics of meropenem in infants and children: a population analysis.
        J Antimicrob Chemother. 1995; 36 (Suppl A): 63-71
        • Pea F
        • Viale P
        • Furlanut M.
        Antimicrobial therapy in critically ill patients: a review of pathophysiological conditions responsible for altered disposition and pharmacokinetic variability.
        Clin Pharmacokinet. 2005; 44: 1009-1034
        • Rodieux F
        • Wilbaux M
        • van den Anker JN
        • Pfister M.
        Effect of Kidney Function on Drug Kinetics and Dosing in Neonates, Infants, and Children.
        Clinical pharmacokinetics. 2015; 54: 1183-1204
        • Schlapbach LJ
        • Straney L
        • Bellomo R
        • MacLaren G
        • Pilcher D.
        Prognostic accuracy of age-adapted SOFA, SIRS, PELOD-2, and qSOFA for in-hospital mortality among children with suspected infection admitted to the intensive care unit.
        Intensive Care Med. 2018; 44: 179-188
        • Schwartz GJ
        • Work DF.
        Measurement and estimation of GFR in children and adolescents.
        Clin J Am Soc Nephrol. 2009; 4: 1832-1843
        • Smith PB
        • Cohen-Wolkowiez M
        • Castro LM
        • Poindexter B
        • Bidegain M
        • Weitkamp JH
        • et al.
        Population pharmacokinetics of meropenem in plasma and cerebrospinal fluid of infants with suspected or complicated intra-abdominal infections.
        Pediatr Infect Dis J. 2011; 30: 844-849
        • van den Anker JN
        • Pokorna P
        • Kinzig-Schippers M
        • Martinkova J
        • de Groot R
        • Drusano GL
        • et al.
        Meropenem pharmacokinetics in the newborn.
        Antimicrob Agents Chemother. 2009; 53: 3871-3879
        • Wang ZM
        • Chen XY
        • Bi J
        • Wang MY
        • Xu BP
        • Tang BH
        • et al.
        Reappraisal of the Optimal Dose of Meropenem in Critically Ill Infants and Children: a Developmental Pharmacokinetic-Pharmacodynamic Analysis.
        Antimicrob Agents Chemother. 2020; 64
        • Yu Z
        • Pang X
        • Wu X
        • Shan C
        • Jiang S.
        Clinical outcomes of prolonged infusion (extended infusion or continuous infusion) versus intermittent bolus of meropenem in severe infection: A meta-analysis.
        PLoS One. 2018; 13e0201667
        • Zheng J
        • Yao Y
        • Han L
        • Xiao Y.
        Renal function and injury in infants and young children with congenital heart disease.
        Pediatr Nephrol. 2013; 28: 99-104