TY - JOUR
AU1 - Matthieu, Boisson,
AU2 - Olivier, Mimoz,
AU3 - Mirza, Hadzic,
AU4 - Sandrine, Marchand,
AU5 - Christophe, Adier,
AU6 - William, Couet,
AU7 - Nicolas, Grégoire,
AB - Abstract Objectives Optimal dosing for nebulized gentamicin is unknown. We compared the pulmonary and systemic pharmacokinetics (PK) of gentamicin following intravenous and nebulized administration in mechanically ventilated patients. Methods Twelve critically ill male patients with ventilator-associated pneumonia received a 30 min intravenous infusion of 8 mg/kg gentamicin , followed 48 h afterwards by the same dose nebulized. Blood samples were collected immediately before and until 24 h after intravenous and nebulized administration; mini-bronchoalveolar lavages (mini-BALs) were performed at 3 and 7 h or 5 and 10 h (six patients each) after each intravenous and nebulized administration. The PK analysis was conducted using a population approach. Results After intravenous administration, concentrations of gentamicin measured in epithelial lining fluid (ELF) were very variable, and overall in the same range of magnitude (from 0.3 to 28 mg/L) as in plasma. After nebulization, gentamicin concentrations were much higher (∼3800-fold) in ELF than in plasma. The average systemic bioavailability of nebulized gentamicin was estimated to be 5%, with considerable inter-individual variability. Compared with intravenous administration, after nebulization the exposure (expressed as AUC) to gentamicin was 276-fold greater in ELF and 18-fold lower in plasma. Conclusions Compared with intravenous administration, nebulization of gentamicin in patients with ventilator-associated pneumonia provides higher pulmonary concentrations and lower systemic concentrations but the inter-individual variability is large. Introduction Ventilator-associated pneumonia (VAP) is the most common complication due to mechanical ventilation and represents nearly half of all healthcare-associated infections, affecting 10%–27% of ventilated patients.1,2 In spite of a clear improvement in outcomes in the last two decades, VAP is still responsible for high mortality, ranging from 9%–13% in recent studies.3 While VAP prognosis is linked to early effective antimicrobial therapy, poor lung tissue penetration of many antibiotics increases the risk of treatment failure.4 Nebulization of antibiotics, by providing high intrapulmonary antibiotic concentrations with limited systemic exposure,5 has shown promising results for VAP therapy in critically ill patients.6,7 Empirically used for decades, nebulized antibiotics are of particular interest for the delivery of hydrophilic antibiotics such as colistin8–10 and aminoglycosides.11 Aminoglycosides, after more than 50 years of use, still retain efficacy against many Gram-negative bacilli and Gram-positive cocci (including Staphylococcus aureus).12 With their narrow therapeutic index13 and poor lung penetration,11,14 aminoglycosides appear to be appropriate antibiotics for nebulization in combination with an intravenous β-lactam or glycopeptide. Gentamicin, which is particularly effective against Gram-positive cocci, is commonly used by nebulization to treat early VAP.15 However, there is a lack of published data concerning the PK of gentamicin after nebulization and there is no consensus regarding the optimal use of this administration route. The efficacy of aminoglycosides has been linked to the ratio of the maximal concentration divided by the MIC (Cmax/MIC) of the pathogen.16,17 On the other hand, nephrotoxicity of gentamicin is correlated with the trough concentration in plasma (Cmin), the range of Cmin that has been reported to be nephrotoxic being 0.5–4 mg/L.18,19 The aim of this study was to compare pulmonary and systemic concentrations of gentamicin following intravenous administration and nebulization of 8 mg/kg gentamicin in critically ill patients with VAP. Patients and methods Study population The study was conducted between October 2015 and November 2016 in 12 adult patients hospitalized in the surgical ICU of the University Hospital of Poitiers, France. Patients were eligible if they were ≥18 years of age and developed VAP due to gentamicin-susceptible pathogens and gentamicin treatment was justified. Patients were excluded if they had received aminoglycosides at any time during the 7 days preceding the study, had a CLCR <60 mL/min, a BMI >40 kg/m2, poor respiratory function (PaO2/FiO2 <150) or a personal or family history of myasthenia. At study onset, the following data were collected: age, sex, weight, height, diagnosis on admission, serum urea, serum creatinine, identification of pathogens, simplified acute physiology score (SAPS II)20 and SOFA score.21 CLCR was estimated according to the Modification of Diet in Renal Disease (MDRD) formula.22 Ethics The study protocol was approved by the local ethics committee (Comité de Protection des Personnes Ouest III, approval number 2014-001665-27) and registered on ClinicalTrials.gov (number NCT02515448). Written informed consent was obtained for all patients from their nearest relatives prior to initiation of the study. Gentamicin administration Treatment was initiated with one 30 min intravenous infusion of 8 mg/kg gentamicin sulphate (Gentamicine Panpharma®, Panpharma, Fougères, France) diluted in saline solution according to French guidelines.23 A second intravenous infusion (same dose) was administered 24 h after the first dose if the plasma concentration of gentamicin was below 0.5 mg/L. At 48 h after the beginning of treatment, i.e. 24 or 48 h after the last intravenous dose, 8 mg/kg gentamicin was nebulized over 30 min using a vibrating mesh nebulizer (Aeroneb Pro®, Aerogen, Galway, Ireland). During aerosol delivery, all patients were sedated and received volume-controlled mechanical ventilation with a tidal volume of 7–8 mL/kg of predicted body weight and respiratory rates of 12–15 cycles/min. If the patient was ventilated with a heat and moisture exchanger, it was switched off and the circuit was changed. The nebulizer was inserted near the Y-piece connector on the inspiratory limb. During and within 1 h after the aerosol delivery, respiratory failure (defined by a decrease of >20% of PaO2/FiO2) or bronchospasm (defined by the use of bronchodilatory therapy) were recorded. Sampling procedures Blood samples Blood samples were collected immediately before and 0.5, 1, 3, 5, 7, 10, 18 and 24 h after the first infused and nebulized gentamicin administration. They were immediately centrifuged at the bedside (3000 g for 10 min at 4°C) and the plasma was stored at −80°C pending assay. Bronchoalveolar lavage fluid samples Mini-bronchoalveolar lavage (BAL) was performed exactly as previously described.9 Briefly, mini-BAL was performed with 16 Fr double sterile catheters (BAL, KimVent®, Kimberly-Clark, Roswell, GA) inserted through the endotracheal tube. Two 20 mL aliquots of saline solution were instilled and then immediately aspirated with a syringe; these two BAL fluid samples were pooled and immediately centrifuged at the bedside (3000 g for 10 min at 4°C). Supernatants were stored at −80°C until analysis. For the first six patients, a mini-BAL was performed at 3 and 7 h after the beginning of the first intravenous infusion and after nebulization. For the last six patients, a mini-BAL was performed at 5 and 10 h after the beginning of the first intravenous infusion and after nebulization. Urine samples Urine samples were collected during the 24 h after the first intravenous infusion and after nebulization. At the end of the 24 h collection, the volume of urine was measured, urine was homogenized and a sample was stored at −80°C until analysis. Gentamicin assay in plasma, BAL fluid and urine A bioanalytical LC-MS/MS method was developed and validated to determine gentamicin concentrations in plasma, BAL and urine samples. The chromatographic analysis was carried out on an Alliance 1295 system (Waters), coupled to a triple quadrupole mass spectrometer, Quattro micro Api (Waters). Chromatographic separation was achieved on a C18 column (Waters XBridge BEH300, 150 mm × 2.1 mm, particle size 5 μM) using a Phenomenex Gemini as a pre-column. Gentamicin was separated using a mobile phase consisting of water and acetonitrile, both with 0.1% formic acid. Sisomicin was used as the internal standard (IS). Drugs were monitored using electrospray ionization operating in positive mode (ESI+) monitoring the transitions 478.2 > 160.1 m/z for gentamicin C1, 450.2 > 160.1 m/z for gentamicin C1A, 464.2 > 160.1 m/z for gentamicin C2 + C2A and 448.2 > 160.1 m/z for the IS. Gentamicin concentration was calculated as the sum of the different transitions (C1 + C1A + C2 + C2A). For each matrix, the method was linear within the concentration range of 0.01–2 mg/L, with coefficients of determination ≥0.991. The dilution of samples with a concentration higher than the range of calibration was validated. The inter- and intra-day precision, the accuracy and the stability of the drug in different matrices were in accordance with the limits established by US FDA and EMA guidelines (15% within the concentration range except 20% at the lower limit of quantification). Urea analysis in plasma and BAL fluid A previously described LC-MS/MS assay was used for urea concentration measurements in BAL fluid samples.8,9 The limit of quantification of the assay was 1 mg/L in BAL fluid. Urea concentrations in plasma were measured by photometric detection using an automatic analyser (Modular automatic analyser; Roche, France; limit of quantification = 30 mg/L). Estimation of gentamicin concentrations in epithelial lining fluid Actual epithelial lining fluid (ELF) concentrations of gentamicin (CELF) were obtained from measured BAL fluid concentrations after correction for dilution,24 according to the following equation: CELF=CBAL(Ureaplasma/UreaBAL) where CBAL corresponds to the gentamicin concentration measured in BAL fluid, and Ureaplasma and UreaBAL correspond to the concentrations of urea determined in plasma and BAL fluid, respectively. Population PK modelling Concentration data were analysed by a population approach with the software package NONMEM, version 7.2.0 (Icon Development Solutions, Ellicott City, MD, USA, 2009). Model fitting was performed with g95 FORTRAN Compiler, version 0.91 (GNU General Public License, GPL), Xpose (xpose.sourceforge.net), PsN, version 4.4.8 (https://uupharmacometrics.github.io/PsN). R, version 3.1.2 (R-project, www.r-project.org) was used for the exploratory analysis, executing NONMEM runs and post-processing of NONMEM output. For the PK analysis, gentamicin concentrations in plasma, ELF and urine were analysed simultaneously. Model development was conducted with the first-order conditional estimation method (FOCE) with interaction. Several PK structural models were compared, including a model with non-renal clearance of gentamicin since urine samples allowed discrimination of non-renal and renal clearances. Between-subject variability (BSV) and inter-occasion variability (IOV) for PK parameters were evaluated using an exponential model. Three distinct residual variabilities were estimated for plasma, ELF and urine concentrations. For each of them, additive, proportional and combined (additive + proportional) models of error were explored. The relationships between gentamicin clearance and CLCR (estimated with the MDRD formula and capped at 130 mL/min as previously25) and between the body weight and V were explored.26 Discrimination between models was based on the inspection of graphical diagnostics and changes in the objective function value (OFV) provided by NONMEM. For a more complicated model to be retained it had to provide a significant improvement over the contending model (P < 0.05 for nested models) and provide parameter values with a good precision of estimation, i.e. with low relative standard error (RSE). In the case of non-nested models, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) values were used to discriminate between models. Model evaluation included graphical analysis of goodness-of-fit plots, RSEs, visual predictive checks (VPCs) and normalized prediction distribution errors (NPDE). The final population PK model was used to simulate 10 000 individual concentration time courses. Simulations incorporated between-patient variability estimated with the final population PK model, but ignored parameter uncertainty. These simulations were used to estimate the distributions of the following secondary PK parameters: plasma Cmax values were obtained from individual simulations; plasma Cmin values were obtained from individual concentrations simulated 24 h after dosing; and plasma AUC0–24 (AUCplasma) and ELF AUC0–24 (AUCELF) were calculated by integration of concentrations over time. Results Patients A total of 12 men were enrolled. Their demographic, clinical and biological data are shown in Table 1. Patient 12 was excluded from the study because of CLCR <60 mL/min during the study. Table 1. Demographic and clinical characteristics of the patients Study start values Patients Age (years) Gender Height (cm) Weight (kg) Diagnosis at admission SOFA score SAPS II score PaO2/FiO2 Albuminaemia (g/L) at study onset CLCR (mL/min) according to MDRD formula 1 46 male 180 80 multiple trauma 4 32 322 21 126 2 56 male 175 100 respiratory failure 7 24 152 30 211 3 32 male 175 70 multiple trauma 9 45 178 23 141 4 19 male 190 93 traumatic brain injury 11 42 172 30 166 5 32 male 168 74 multiple trauma 8 34 213 28 277 6 49 male 177 100 brain injury 6 31 153 34 159 7 65 male 178 100 traumatic brain injury 10 44 160 29 88 8 19 male 160 52 multiple trauma 8 53 280 31 203 9 35 male 190 99 whiplash injury 4 19 151 32 120 10 39 male 180 82 multiple trauma 6 27 208 25 214 11 44 male 176 55 respiratory failure 10 47 164 25 88 Overall (mean ± SD) 40 ± 14 178 ± 8 83 ± 17 8 ± 2 37 ± 11 192 ± 55 28 ± 3 156 ± 60 Study start values Patients Age (years) Gender Height (cm) Weight (kg) Diagnosis at admission SOFA score SAPS II score PaO2/FiO2 Albuminaemia (g/L) at study onset CLCR (mL/min) according to MDRD formula 1 46 male 180 80 multiple trauma 4 32 322 21 126 2 56 male 175 100 respiratory failure 7 24 152 30 211 3 32 male 175 70 multiple trauma 9 45 178 23 141 4 19 male 190 93 traumatic brain injury 11 42 172 30 166 5 32 male 168 74 multiple trauma 8 34 213 28 277 6 49 male 177 100 brain injury 6 31 153 34 159 7 65 male 178 100 traumatic brain injury 10 44 160 29 88 8 19 male 160 52 multiple trauma 8 53 280 31 203 9 35 male 190 99 whiplash injury 4 19 151 32 120 10 39 male 180 82 multiple trauma 6 27 208 25 214 11 44 male 176 55 respiratory failure 10 47 164 25 88 Overall (mean ± SD) 40 ± 14 178 ± 8 83 ± 17 8 ± 2 37 ± 11 192 ± 55 28 ± 3 156 ± 60 Pathogens cultured were the following Gram-positive organisms: S. aureus, Streptococcus anginosus, Streptococcus mitis and Streptococcus pneumoniae (with MICs from 0.5–16 mg/L) and the following Gram-negative organisms: Acinetobacter baumannii, Citrobacter koseri, Enterobacter aerogenes, Enterobacter cloacae and Escherichia coli (with MICs <1 mg/L). Table 1. Demographic and clinical characteristics of the patients Study start values Patients Age (years) Gender Height (cm) Weight (kg) Diagnosis at admission SOFA score SAPS II score PaO2/FiO2 Albuminaemia (g/L) at study onset CLCR (mL/min) according to MDRD formula 1 46 male 180 80 multiple trauma 4 32 322 21 126 2 56 male 175 100 respiratory failure 7 24 152 30 211 3 32 male 175 70 multiple trauma 9 45 178 23 141 4 19 male 190 93 traumatic brain injury 11 42 172 30 166 5 32 male 168 74 multiple trauma 8 34 213 28 277 6 49 male 177 100 brain injury 6 31 153 34 159 7 65 male 178 100 traumatic brain injury 10 44 160 29 88 8 19 male 160 52 multiple trauma 8 53 280 31 203 9 35 male 190 99 whiplash injury 4 19 151 32 120 10 39 male 180 82 multiple trauma 6 27 208 25 214 11 44 male 176 55 respiratory failure 10 47 164 25 88 Overall (mean ± SD) 40 ± 14 178 ± 8 83 ± 17 8 ± 2 37 ± 11 192 ± 55 28 ± 3 156 ± 60 Study start values Patients Age (years) Gender Height (cm) Weight (kg) Diagnosis at admission SOFA score SAPS II score PaO2/FiO2 Albuminaemia (g/L) at study onset CLCR (mL/min) according to MDRD formula 1 46 male 180 80 multiple trauma 4 32 322 21 126 2 56 male 175 100 respiratory failure 7 24 152 30 211 3 32 male 175 70 multiple trauma 9 45 178 23 141 4 19 male 190 93 traumatic brain injury 11 42 172 30 166 5 32 male 168 74 multiple trauma 8 34 213 28 277 6 49 male 177 100 brain injury 6 31 153 34 159 7 65 male 178 100 traumatic brain injury 10 44 160 29 88 8 19 male 160 52 multiple trauma 8 53 280 31 203 9 35 male 190 99 whiplash injury 4 19 151 32 120 10 39 male 180 82 multiple trauma 6 27 208 25 214 11 44 male 176 55 respiratory failure 10 47 164 25 88 Overall (mean ± SD) 40 ± 14 178 ± 8 83 ± 17 8 ± 2 37 ± 11 192 ± 55 28 ± 3 156 ± 60 Pathogens cultured were the following Gram-positive organisms: S. aureus, Streptococcus anginosus, Streptococcus mitis and Streptococcus pneumoniae (with MICs from 0.5–16 mg/L) and the following Gram-negative organisms: Acinetobacter baumannii, Citrobacter koseri, Enterobacter aerogenes, Enterobacter cloacae and Escherichia coli (with MICs <1 mg/L). Plasma and pulmonary PK Five values of gentamicin concentration in ELF were removed from the analysis because the UreaBAL was lower than the limit of quantification. Plasma and ELF concentrations after administration of intravenous and nebulized 8 mg/kg gentamicin are shown in Table 2. Table 2. Plasma and ELF concentrations after administration of 8 mg/kg gentamicin, intravenous or nebulized Time after administration (h) Intravenous infusion Nebulization plasma ELF plasma ELF 3 7.5 ± 1.4 2.9 ± 1.7 0.6 ± 0.7 4359 ± 10 255 5 6.2 ± 2.5 6.5 ± 4.9 0.8 ± 0.6 3422 ± 4028 7 3.0 ± 1.4 2.7 ± 2.0 0.4 ± 0.5 2828 ± 4338 10 2.5 ± 1.6 9.7 ± 12.4 0.5 ± 0.2 1817 ± 1626 Time after administration (h) Intravenous infusion Nebulization plasma ELF plasma ELF 3 7.5 ± 1.4 2.9 ± 1.7 0.6 ± 0.7 4359 ± 10 255 5 6.2 ± 2.5 6.5 ± 4.9 0.8 ± 0.6 3422 ± 4028 7 3.0 ± 1.4 2.7 ± 2.0 0.4 ± 0.5 2828 ± 4338 10 2.5 ± 1.6 9.7 ± 12.4 0.5 ± 0.2 1817 ± 1626 Values are expressed as mean ± SD (mg/L). Table 2. Plasma and ELF concentrations after administration of 8 mg/kg gentamicin, intravenous or nebulized Time after administration (h) Intravenous infusion Nebulization plasma ELF plasma ELF 3 7.5 ± 1.4 2.9 ± 1.7 0.6 ± 0.7 4359 ± 10 255 5 6.2 ± 2.5 6.5 ± 4.9 0.8 ± 0.6 3422 ± 4028 7 3.0 ± 1.4 2.7 ± 2.0 0.4 ± 0.5 2828 ± 4338 10 2.5 ± 1.6 9.7 ± 12.4 0.5 ± 0.2 1817 ± 1626 Time after administration (h) Intravenous infusion Nebulization plasma ELF plasma ELF 3 7.5 ± 1.4 2.9 ± 1.7 0.6 ± 0.7 4359 ± 10 255 5 6.2 ± 2.5 6.5 ± 4.9 0.8 ± 0.6 3422 ± 4028 7 3.0 ± 1.4 2.7 ± 2.0 0.4 ± 0.5 2828 ± 4338 10 2.5 ± 1.6 9.7 ± 12.4 0.5 ± 0.2 1817 ± 1626 Values are expressed as mean ± SD (mg/L). The non-renal clearance of gentamicin was negligible, and thus the total clearance was considered equal to the renal clearance. Accordingly, from concentrations measured in urine, 24 h after intravenous infusion, most of the dose (82%) of gentamicin was excreted in urine. The structural PK model used to fit the data is presented in Figure 1. Gentamicin was cleared from a central compartment (of volume Vc) by CLR, distributed towards a peripheral compartment with an intercompartmental clearance (Qplasma) and towards ELF with an intercompartmental clearance (CLdif). CLR was related to the CLCR of patients following the equation: CLR=CLR, TV×CLCR110β in which CLR, TV is the typical value of CLR for a patient with a CLCR of 110 mL/min and β is the exponent of the power function. V was not significantly related to body weight. In the lung, gentamicin was distributed between an ELF compartment (of volume VELF) and a peripheral compartment (of volume Vplung); Qlung refers to the intercompartmental clearance in the lung. After aerosol delivery, the fraction of the gentamicin dose that reached the ELF was defined as FAERO. Residual variabilities for plasma, ELF and urine concentrations were proportional. The addition of compartments, whether in the systemic or in the lung part of the model, the addition of a non-renal clearance, or the addition of clearances for active transport between plasma and lung did not significantly improve the data fit. Figure 1. View largeDownload slide Structural PK model. VELF, volume of distribution of ELF; Vplung, volume of distribution of peripheral compartment in the lung; Qlung, distribution clearance in the lung; FAERO, fraction of the aerosol dose that reached the systemic circulation; CLdif, diffusion clearance between plasma and ELF; Vc, volume of distribution of the central compartment; Vpplasma, volume of distribution of the systemic peripheral compartment; Qplasma, systemic distribution clearance. Figure 1. View largeDownload slide Structural PK model. VELF, volume of distribution of ELF; Vplung, volume of distribution of peripheral compartment in the lung; Qlung, distribution clearance in the lung; FAERO, fraction of the aerosol dose that reached the systemic circulation; CLdif, diffusion clearance between plasma and ELF; Vc, volume of distribution of the central compartment; Vpplasma, volume of distribution of the systemic peripheral compartment; Qplasma, systemic distribution clearance. PK modelling allowed a proper description of gentamicin concentrations in plasma, ELF and urine after intravenous administration and nebulization. The VPC showed that the model fitted the data without major bias (Figure 2, data not shown for urine). The observed concentrations were evenly scattered around the typical profile and ∼6% of the observed data were outside the 90% predicted confidence intervals. Distributions of NPDE were close to Gaussian distribution (data not shown). The parameter estimates are presented in Tables 3 and 4. Table 3. Estimated population PK parameters of gentamicin for the final model based on the data from 11 patients Description Parameters Units Typical value (RSE%) IIV CV% (RSE%) IOV CV% (RSE%) Lung  fraction of the dose that reaches systemic circulation FAERO 0.05 (31) 155 (49) —  clearance of diffusion between ELF and plasma CLdif mL/h 1.4 (44) 57 (68)  ELF volume VELF mL 2.1 (65) 135 (72) —  peripheral volume in lung Vplung mL 9.5 (63) 127 (53)  distribution clearance in lung Qlung mL/h 3.6 (56) Plasma  volume of central compartment VC L 20.3 (9) 22 (57) —  typical value of total (= renal) clearance CLR, TV mL/min 95.0 (11) 26 (34) —  exponent of the power model for CLR β 1.6 (45)  volume of peripheral compartment Vpplasma L 9.6 (17) 26 (87)  distribution clearance in plasma Qplasma mL/min 19.3 (24) Residual error, proportional (%)  plasma 20 (21) —  ELF 105 (57) —  urine 29 (30) Description Parameters Units Typical value (RSE%) IIV CV% (RSE%) IOV CV% (RSE%) Lung  fraction of the dose that reaches systemic circulation FAERO 0.05 (31) 155 (49) —  clearance of diffusion between ELF and plasma CLdif mL/h 1.4 (44) 57 (68)  ELF volume VELF mL 2.1 (65) 135 (72) —  peripheral volume in lung Vplung mL 9.5 (63) 127 (53)  distribution clearance in lung Qlung mL/h 3.6 (56) Plasma  volume of central compartment VC L 20.3 (9) 22 (57) —  typical value of total (= renal) clearance CLR, TV mL/min 95.0 (11) 26 (34) —  exponent of the power model for CLR β 1.6 (45)  volume of peripheral compartment Vpplasma L 9.6 (17) 26 (87)  distribution clearance in plasma Qplasma mL/min 19.3 (24) Residual error, proportional (%)  plasma 20 (21) —  ELF 105 (57) —  urine 29 (30) Table 3. Estimated population PK parameters of gentamicin for the final model based on the data from 11 patients Description Parameters Units Typical value (RSE%) IIV CV% (RSE%) IOV CV% (RSE%) Lung  fraction of the dose that reaches systemic circulation FAERO 0.05 (31) 155 (49) —  clearance of diffusion between ELF and plasma CLdif mL/h 1.4 (44) 57 (68)  ELF volume VELF mL 2.1 (65) 135 (72) —  peripheral volume in lung Vplung mL 9.5 (63) 127 (53)  distribution clearance in lung Qlung mL/h 3.6 (56) Plasma  volume of central compartment VC L 20.3 (9) 22 (57) —  typical value of total (= renal) clearance CLR, TV mL/min 95.0 (11) 26 (34) —  exponent of the power model for CLR β 1.6 (45)  volume of peripheral compartment Vpplasma L 9.6 (17) 26 (87)  distribution clearance in plasma Qplasma mL/min 19.3 (24) Residual error, proportional (%)  plasma 20 (21) —  ELF 105 (57) —  urine 29 (30) Description Parameters Units Typical value (RSE%) IIV CV% (RSE%) IOV CV% (RSE%) Lung  fraction of the dose that reaches systemic circulation FAERO 0.05 (31) 155 (49) —  clearance of diffusion between ELF and plasma CLdif mL/h 1.4 (44) 57 (68)  ELF volume VELF mL 2.1 (65) 135 (72) —  peripheral volume in lung Vplung mL 9.5 (63) 127 (53)  distribution clearance in lung Qlung mL/h 3.6 (56) Plasma  volume of central compartment VC L 20.3 (9) 22 (57) —  typical value of total (= renal) clearance CLR, TV mL/min 95.0 (11) 26 (34) —  exponent of the power model for CLR β 1.6 (45)  volume of peripheral compartment Vpplasma L 9.6 (17) 26 (87)  distribution clearance in plasma Qplasma mL/min 19.3 (24) Residual error, proportional (%)  plasma 20 (21) —  ELF 105 (57) —  urine 29 (30) Table 4. Plasma and ELF exposure after administration of 8 mg/kg gentamicin, intravenous or nebulized Description Parameters Units Intravenous Nebulized Maximum concentration in plasma Cmax mg/L 29.6 (29) 0.50 (114) Minimum concentration in plasma Cmin mg/L 0.24 (115) 0.07 (130) Probability that Cmin >0.5 mg/L P (Cmin >0.5) — 23% 3% AUC0–24 in plasma AUCplasma mg·h/L 95.4 (40) 5.4 (111) AUC0–24 in ELF AUCELF mg·h/L 73.9 (46) 20 367 (131) Description Parameters Units Intravenous Nebulized Maximum concentration in plasma Cmax mg/L 29.6 (29) 0.50 (114) Minimum concentration in plasma Cmin mg/L 0.24 (115) 0.07 (130) Probability that Cmin >0.5 mg/L P (Cmin >0.5) — 23% 3% AUC0–24 in plasma AUCplasma mg·h/L 95.4 (40) 5.4 (111) AUC0–24 in ELF AUCELF mg·h/L 73.9 (46) 20 367 (131) Parameters are expressed as median (CV%). Table 4. Plasma and ELF exposure after administration of 8 mg/kg gentamicin, intravenous or nebulized Description Parameters Units Intravenous Nebulized Maximum concentration in plasma Cmax mg/L 29.6 (29) 0.50 (114) Minimum concentration in plasma Cmin mg/L 0.24 (115) 0.07 (130) Probability that Cmin >0.5 mg/L P (Cmin >0.5) — 23% 3% AUC0–24 in plasma AUCplasma mg·h/L 95.4 (40) 5.4 (111) AUC0–24 in ELF AUCELF mg·h/L 73.9 (46) 20 367 (131) Description Parameters Units Intravenous Nebulized Maximum concentration in plasma Cmax mg/L 29.6 (29) 0.50 (114) Minimum concentration in plasma Cmin mg/L 0.24 (115) 0.07 (130) Probability that Cmin >0.5 mg/L P (Cmin >0.5) — 23% 3% AUC0–24 in plasma AUCplasma mg·h/L 95.4 (40) 5.4 (111) AUC0–24 in ELF AUCELF mg·h/L 73.9 (46) 20 367 (131) Parameters are expressed as median (CV%). Figure 2. View largeDownload slide Observed gentamicin concentrations in ELF (top) and plasma (bottom) with model predictions for medians (solid decay curve) and for 90% prediction intervals (shaded areas) after 8 mg/kg doses administered as intravenous infusions (right) or nebulizations (left). Figure 2. View largeDownload slide Observed gentamicin concentrations in ELF (top) and plasma (bottom) with model predictions for medians (solid decay curve) and for 90% prediction intervals (shaded areas) after 8 mg/kg doses administered as intravenous infusions (right) or nebulizations (left). After intravenous administration the PK of gentamicin in plasma was biphasic with typical t½ of the first and second phases of 1.9 and 7.5 h respectively. The distribution clearance between plasma and ELF (CLdif = 1.4 mL/h) was much lower than between plasma and the peripheral compartment (Qplasma = 19.3 mL/min). The median Cmax was predicted to be 29.6 mg/L and the median Cmin 0.24 mg/L, with 23% of patients predicted to have Cmin >0.5 mg/L. Concentrations measured in ELF after intravenous administration were in the same range as plasma concentrations (from 0.3 to 28 mg/L) and very variable between patients (Figure 2). Following gentamicin nebulization, gentamicin concentrations (expressed as AUC) were much higher (∼4000-fold) in ELF than in plasma (Figure 2 and Table 4). The systemic bioavailability of gentamicin was estimated to be 5%. Accordingly, from concentrations measured in urine, 24 h after nebulization, on average, 5.6% of the gentamicin dose was retrieved in urine. The PK of gentamicin in ELF was biphasic, with t½ of the first and second phases of 0.3 and 7.3 h respectively. The peak in plasma occurred between 0.33 and 3 h after the start of nebulization, the typical Cmax and Cmin in plasma were predicted to be 0.50 and 0.07 mg/L respectively, with 3% of patients predicted to have Cmin >0.5 mg/L. The large variability of gentamicin concentrations observed in ELF after nebulization (from 1.6 to 25 290 mg/L) and after intravenous administration (from 0.3 to 28 mg/L) was characterized in the model by large inter-individual variabilities (IIVs) for intrapulmonary parameters (FAERO, VELF and Vplung, ranging between 127% and 155%) and large residual variability (105%). Compared with intravenous administration, after nebulization the exposure to gentamicin, as assessed by AUC, was 276-fold greater in ELF and 18-fold lower in plasma, on average (Table 4). Safety During and after nebulization, no respiratory failure or bronchospasm occurred. Discussion The systemic and the lung parts of the PK model were both bicompartmental (Figure 1). In our population with preserved renal function, the typical renal clearance of gentamicin was 95 mL/min. V at steady-state was 29.9 L and the terminal t½ was 7.5 h. These results were in agreement with those previously reported.17 The diffusion of gentamicin between ELF and plasma was shown to have a unique distribution clearance, meaning that after intravenous administration AUCs of total concentrations in plasma and ELF were equal. This result has to be confirmed because of the high variability of gentamicin concentrations in ELF and of the early termination of BAL sampling (10 h). This is in accordance with passive diffusion of gentamicin if we assume that protein binding in plasma and tissue was similar. High protein binding of gentamicin within ELF could be responsible for high concentrations in ELF. Protein binding of gentamicin in plasma is known to be low (<30%) but binding in ELF is unknown.27,28 Only one study had been previously conducted to investigate gentamicin concentrations within ELF. Two hours after the start of a 30 min intravenous administration of 3.5 mg/kg gentamicin, gentamicin concentrations were about three times lower in ELF than in plasma, which is in accordance with our results at the same time (predicted ratio of 2.5).14 However, after intravenous administration the peak of gentamicin in ELF was delayed and decreased compared with that in plasma, confirming that the rate of diffusion of gentamicin within ELF was slow. Therefore, the ratio measured at 2 h was lower than those observed later (Figure 2), indicating that a single measure in tissue might not be representative of tissue distribution over time.29 It is notable that we ended BAL sampling 10 h after dosing; subsequent predictions of gentamicin concentrations should be considered with caution. After nebulization, concentrations of gentamicin in ELF (as assessed by AUC) were ∼300-fold greater, on average, than after intravenous administration. Gentamicin concentrations in ELF were associated with high residual variability (105%). Such high residual variability had already been described in other studies, owing to BAL sampling, assay procedures and correction of BAL concentrations by the urea method.9,10,30,31 It is of note that in order to reduce the residual variability, in this study, sampling of ELF was done by mini-BAL, which is a reliable method compared with bronchoscopic BAL.32 Moreover, the extremity of the catheter was protected, which reduced the contamination of the sample by gentamicin deposited on the trachea and stem bronchi. Moreover, it can be noted that after nebulization the variability of gentamicin concentrations in ELF was greater than after intravenous administration because of large IIVs of pulmonary parameters (Table 3). According to the PK model, the high concentrations measured in ELF were due to relatively slow passage of gentamicin from ELF to plasma (Qlung = 3.6 mL/h). This systemic passage was driven by the distribution of gentamicin within the lung and was biphasic, with a first phase t½ of 0.3 h, explaining the early peak in plasma, and a second phase t½ of 7.3 h. The poor permeation of gentamicin through physiological membranes is in accordance with its physicochemical properties since gentamicin is highly hydrophilic and polycationic at physiological pH (Log D = −12.4 and net charge of 5 at pH = 7.4) with a high polar surface area (199.7 Å2).33 The very high concentrations of gentamicin measured in ELF achieved with nebulization makes this a route of administration which attracts great interest. However, the level of gentamicin concentration in ELF necessary to be effective against pulmonary infections is unknown. The efficacy of aminoglycosides has been reported to be correlated with Cmax/MIC values greater than 7 to 10.16,17 However, these targets refer to plasma peak concentrations and their application to pulmonary concentrations may fail to predict aminoglycoside efficacy.34 Indeed, plasma concentrations might poorly reflect activity of gentamicin in the lung because biofilms can slow the diffusion of aminoglycosides and because airway surface liquids contain proteins, such as mucin, to which aminoglycosides can bind,35,36 resulting in a dramatic decrease of their activity.37–39 Concentrations needed in the lung to be effective might therefore be higher than in other sites of infection, thus reinforcing the interest in reaching high concentrations with nebulization. We report for the first time the bioavailability of gentamicin after nebulization (FAERO). On average, it was estimated to be 5%, which is consistent with the low bioavailabilities reported after nebulization of colistin or amikacin during mechanical ventilation.9,40 The IIV of FAERO was very large [coefficient of variation (CV) = 155%, 95% prediction interval = 0.7%–32.3%], resulting in a very variable AUC of gentamicin in plasma after nebulization. However, the aerosol delivery was standardized (sedation, ventilator parameters, ventilator circuit and settings) and some individual characteristics must be responsible for this large IIV. In a porcine model of pneumonia, it has been shown that pulmonary diffusion and systemic bioavailability of nebulized amikacin was influenced by experimental conditions, the severity of pneumonia and by the lung aeration.5,41,42 The small number of patients enrolled limited the possibility of exploring this issue in more detail. Moreover, owing to the specific study population (exclusively composed of men and mainly trauma patients), the extension of our results to all critically ill patients should be done with caution. Because of the low FAERO, plasma concentrations were much lower after nebulization of 8 mg/kg gentamicin than after intravenous administration. After nebulization, the typical Cmin in plasma was 0.07 mg/L (0.24 mg/L after intravenous) and only 3% (23% after intravenous) of patients were predicted to have Cmin >0.5 mg/L, which is usually considered a predictive limit value of the systemic toxicity of gentamicin. It is noteworthy that the nebulization was well-tolerated in every patient. However, the number of patients was not sufficient for an appropriate evaluation of safety and larger studies are warranted to address this issue. Conclusions Compared with intravenous administration, nebulization of gentamicin in patients with VAP provides higher pulmonary concentrations and lower systemic concentrations, but the IIV is large. Funding This work was supported by internal research funds of University Hospital of Poitiers and University of Poitiers. Transparency declarations None to declare. References 1 Vincent JL , Bihari DJ , Suter PM et al. The prevalence of nosocomial infection in intensive care units in Europe. Results of the European Prevalence of Infection in Intensive Care (EPIC) Study. EPIC International Advisory Committee . JAMA 1995 ; 274 : 639 – 44 . Google Scholar Crossref Search ADS PubMed 2 Metersky ML , Wang Y , Klompas M et al. Trend in ventilator-associated pneumonia rates between 2005 and 2013 . JAMA 2016 ; 316 : 2427 – 9 . Google Scholar Crossref Search ADS PubMed 3 Melsen WG , Rovers MM , Groenwold RH et al. Attributable mortality of ventilator-associated pneumonia: a meta-analysis of individual patient data from randomised prevention studies . Lancet Infect Dis 2013 ; 13 : 665 – 71 . Google Scholar Crossref Search ADS PubMed 4 Rodvold KA , George JM , Yoo L. Penetration of anti-infective agents into pulmonary epithelial lining fluid: focus on antibacterial agents . Clin Pharmacokinet 2011 ; 50 : 637 – 64 . Google Scholar Crossref Search ADS PubMed 5 Goldstein I , Wallet F , Robert J et al. Lung tissue concentrations of nebulized amikacin during mechanical ventilation in piglets with healthy lungs . Am J Respir Crit Care Med 2002 ; 165 : 171 – 5 . Google Scholar Crossref Search ADS PubMed 6 Lu Q , Luo R , Bodin L et al. Efficacy of high-dose nebulized colistin in ventilator-associated pneumonia caused by multidrug-resistant Pseudomonas aeruginosa and Acinetobacter baumannii . Anesthesiology 2012 ; 117 : 1335 – 47 . Google Scholar Crossref Search ADS PubMed 7 Luyt C-E , Bréchot N , Combes A et al. Delivering antibiotics to the lungs of patients with ventilator-associated pneumonia: an update . Expert Rev Anti Infect Ther 2013 ; 11 : 511 – 21 . Google Scholar Crossref Search ADS PubMed 8 Gontijo AV , Grégoire N , Lamarche I et al. Biopharmaceutical characterization of nebulized antimicrobial agents in rats: 2. Colistin . Antimicrob Agents Chemother 2014 ; 58 : 3950 – 6 . Google Scholar Crossref Search ADS PubMed 9 Boisson M , Jacobs M , Grégoire N et al. Comparison of intrapulmonary and systemic pharmacokinetics of colistin methanesulfonate (CMS) and colistin after aerosol delivery and intravenous administration of CMS in critically ill patients . Antimicrob Agents Chemother 2014 ; 58 : 7331 – 9 . Google Scholar Crossref Search ADS PubMed 10 Boisson M , Grégoire N , Cormier M et al. Pharmacokinetics of nebulized colistin methanesulfonate in critically ill patients . J Antimicrob Chemother 2017 ; 72 : 2607 – 12 . Google Scholar Crossref Search ADS PubMed 11 Marchand S , Grégoire N , Brillault J et al. Biopharmaceutical characterization of nebulized antimicrobial agents in rats: 3. Tobramycin . Antimicrob Agents Chemother 2015 ; 59 : 6646 – 7 . Google Scholar Crossref Search ADS PubMed 12 EUCAST . Rationale for the EUCAST Clinical Breakpoints—Gentamicin. 2009 . http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/Rationale_documents/Gentamicin_rationale_1.2_0906.pdf. 13 Begg EJ , Barclay ML. Aminoglycosides—50 years on . Br J Clin Pharmacol 1995 ; 39 : 597 – 603 . Google Scholar Crossref Search ADS PubMed 14 Panidis D , Markantonis SL , Boutzouka E et al. Penetration of gentamicin into the alveolar lining fluid of critically ill patients with ventilator-associated pneumonia . Chest 2005 ; 128 : 545 – 52 . Google Scholar Crossref Search ADS PubMed 15 Ehrmann S , Roche-Campo F , Bodet-Contentin L et al. Aerosol therapy in intensive and intermediate care units: prospective observation of 2808 critically ill patients . Intensive Care Med 2016 ; 42 : 192 – 201 . Google Scholar Crossref Search ADS PubMed 16 Moore RD , Lietman PS , Smith CR. Clinical response to aminoglycoside therapy: importance of the ratio of peak concentration to minimal inhibitory concentration . J Infect Dis 1987 ; 155 : 93 – 9 . Google Scholar Crossref Search ADS PubMed 17 Llanos-Paez CC , Hennig S , Staatz CE. Population pharmacokinetic modelling, Monte Carlo simulation and semi-mechanistic pharmacodynamic modelling as tools to personalize gentamicin therapy . J Antimicrob Chemother 2017 ; 72 : 639 – 67 . Google Scholar PubMed 18 De Waele JJ , De Neve N. Aminoglycosides for life-threatening infections: a plea for an individualized approach using intensive therapeutic drug monitoring . Minerva Anestesiol 2014 ; 80 : 1135 – 42 . Google Scholar PubMed 19 Destache CJ. Aminoglycoside-induced nephrotoxicity—a focus on monitoring: a review of literature . J Pharm Pract 2014 ; 27 : 562 – 6 . Google Scholar Crossref Search ADS PubMed 20 Le Gall JR , Lemeshow S , Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study . JAMA 1993 ; 270 : 2957 – 63 . Google Scholar Crossref Search ADS PubMed 21 Vincent JL , Moreno R , Takala J et al. ; on behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine . The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure . Intensive Care Med 1996 ; 22 : 707 – 10 . Google Scholar Crossref Search ADS PubMed 22 Kang YS , Han KH , Han SY et al. Characteristics of population with normal serum creatinine impaired renal function and: the validation of a MDRD formula in a healthy general population . Clin Nephrol 2005 ; 63 : 258 – 66 . Google Scholar Crossref Search ADS PubMed 23 ANSM . Update on good use of injectable aminoglycosides, gentamycin, tobramycin, netilmycin, amikacin. Pharmacological properties, indications, dosage, and mode of administration, treatment monitoring . Med Mal Infect 2012 ; 42 : 301 – 8 . Crossref Search ADS PubMed 24 Marchand S , Gobin P , Brillault J et al. Aerosol therapy with colistin methanesulfonate: a biopharmaceutical issue illustrated in rats . Antimicrob Agents Chemother 2010 ; 54 : 3702 – 7 . Google Scholar Crossref Search ADS PubMed 25 Grégoire N , Mimoz O , Mégarbane B et al. New colistin population pharmacokinetic data in critically ill patients suggesting an alternative loading dose rationale . Antimicrob Agents Chemother 2014 ; 58 : 7324 – 30 . Google Scholar Crossref Search ADS PubMed 26 Hodiamont CJ , Juffermans NP , Bouman CSC et al. Determinants of gentamicin concentrations in critically ill patients: a population pharmacokinetic analysis . Int J Antimicrob Agents 2017 ; 49 : 204 – 11 . Google Scholar Crossref Search ADS PubMed 27 Gyselynck AM , Forrey A , Cutler R. Pharmacokinetics of gentamicin: distribution and plasma and renal clearance . J Infect Dis 1971 ; 124 Suppl: S70 – 6 . Google Scholar Crossref Search ADS PubMed 28 Bailey DN , Briggs JR. Gentamicin and tobramycin binding to human serum in vitro . J Anal Toxicol 2004 ; 28 : 187 – 9 . Google Scholar Crossref Search ADS PubMed 29 Mouton JW , Theuretzbacher U , Craig WA et al. Tissue concentrations: do we ever learn? J Antimicrob Chemother 2008 ; 61 : 235 – 7 . Google Scholar Crossref Search ADS PubMed 30 Kiem S , Schentag JJ. Interpretation of antibiotic concentration ratios measured in epithelial lining fluid . Antimicrob Agents Chemother 2008 ; 52 : 24 – 36 . Google Scholar Crossref Search ADS PubMed 31 Husain S , Capitano B , Corcoran T et al. Intrapulmonary disposition of amphotericin B after aerosolized delivery of amphotericin B lipid complex (Abelcet; ABLC) in lung transplant recipients . Transplantation 2010 ; 90 : 1215 – 9 . Google Scholar Crossref Search ADS PubMed 32 Boselli E , Breilh D , Djabarouti S et al. Reliability of mini-bronchoalveolar lavage for the measurement of epithelial lining fluid concentrations of tobramycin in critically ill patients . Intensive Care Med 2007 ; 33 : 1519 – 23 . Google Scholar Crossref Search ADS PubMed 33 Rottbøll LAH , Friis C. Penetration of antimicrobials to pulmonary epithelial lining fluid and muscle and impact of drug physicochemical properties determined by microdialysis . J Pharmacol Toxicol Methods 2016 ; 78 : 58 – 65 . Google Scholar Crossref Search ADS PubMed 34 Fish DN , Kiser TH. Correlation of pharmacokinetic/pharmacodynamic-derived predictions of antibiotic efficacy with clinical outcomes in severely ill patients with Pseudomonas aeruginosa pneumonia . Pharmacotherapy 2013 ; 33 : 1022 – 34 . Google Scholar Crossref Search ADS PubMed 35 Cross CE , van der Vliet A , O’Neill CA et al. Oxidants, antioxidants, and respiratory tract lining fluids . Environ Health Perspect 1994 ; 102 Suppl 10: 185 – 91 . Google Scholar PubMed 36 Ng AW , Bidani A , Heming TA. Innate host defense of the lung: effects of lung-lining fluid pH . Lung 2004 ; 182 : 297 – 317 . Google Scholar Crossref Search ADS PubMed 37 Ramphal R , Lhermitte M , Filliat M et al. The binding of anti-pseudomonal antibiotics to macromolecules from cystic fibrosis sputum . J Antimicrob Chemother 1988 ; 22 : 483 – 90 . Google Scholar Crossref Search ADS PubMed 38 Hunt BE , Weber A , Berger A et al. Macromolecular mechanisms of sputum inhibition of tobramycin activity . Antimicrob Agents Chemother 1995 ; 39 : 34 – 9 . Google Scholar Crossref Search ADS PubMed 39 Huang JX , Blaskovich MAT , Pelingon R et al. Mucin binding reduces colistin antimicrobial activity . Antimicrob Agents Chemother 2015 ; 59 : 5925 – 31 . Google Scholar Crossref Search ADS PubMed 40 Petitcollin A , Dequin P-F , Darrouzain F et al. Pharmacokinetics of high-dose nebulized amikacin in ventilated critically ill patients . J Antimicrob Chemother 2016 ; 71 : 3482 – 6 . Google Scholar Crossref Search ADS PubMed 41 Elman M , Goldstein I , Marquette C-H et al. Influence of lung aeration on pulmonary concentrations of nebulized and intravenous amikacin in ventilated piglets with severe bronchopneumonia . Anesthesiology 2002 ; 97 : 199 – 206 . Google Scholar Crossref Search ADS PubMed 42 Goldstein I , Wallet F , Nicolas-Robin A et al. Lung deposition and efficiency of nebulized amikacin during Escherichia coli pneumonia in ventilated piglets . Am J Respir Crit Care Med 2002 ; 166 : 1375 – 81 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Pharmacokinetics of intravenous and nebulized gentamicin in critically ill patients
JF - Journal of Antimicrobial Chemotherapy
DO - 10.1093/jac/dky239
DA - 2018-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/pharmacokinetics-of-intravenous-and-nebulized-gentamicin-in-critically-rYS65pbVfN
SP - 2830
VL - 73
IS - 10
DP - DeepDyve
ER -