- Email: [email protected]
Sevoflurane for procedural sedation in critically ill patients: A pharmacokinetic comparative study between burn and non-burn patients
Accepted Manuscript Title: Sevoflurane For Procedural Sedation In Critically Ill Patients: A Pharmacokinetic Comparative Study Between Burn And Non-Burn Patients Authors: Sebastien Perbet Daniel Bourdeaux Alexandre Lenoire Claire Biboulet Bruno Pereira Malha Sadoune Benoit Plaud Jean-Marie Launay Jean-Etienne Bazin Valerie Sautou Alexandre Mebazaa Pascal Houze Jean-Michel Constantin Matthieu Legrand, for the PRONOBURN group PII: DOI: Reference:
S2352-5568(17)30244-8 https://doi.org/doi:10.1016/j.accpm.2018.02.001 ACCPM 344
To appear in: Received date: Revised date: Accepted date:
2-10-2017 6-2-2018 7-2-2018
Please cite this article as: Sebastien PerbetDaniel BourdeauxAlexandre LenoireClaire BibouletBruno PereiraMalha SadouneBenoit PlaudJean-Marie LaunayJean-Etienne BazinValerie SautouAlexandre MebazaaPascal HouzeJean-Michel ConstantinMatthieu Legrandfor the PRONOBURN group, Sevoflurane For Procedural Sedation In Critically Ill Patients: A Pharmacokinetic Comparative Study Between Burn And Non-Burn Patients (2018), https://doi.org/10.1016/j.accpm.2018.02.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
SEVOFLURANE FOR PROCEDURAL SEDATION IN CRITICALLY ILL PATIENTS: A PHARMACOKINETIC COMPARATIVE STUDY BETWEEN BURN AND NON-BURN PATIENTS
ip t
Sebastien Perbet M.D. 1,2, Daniel Bourdeaux PharmD.,PhD. 3,4, Alexandre Lenoire M.D. 5, Claire Biboulet M.D. 1, Bruno Pereira PhD. 6, Malha Sadoune 7, Benoit Plaud M.D., PhD. 5,7,8,
cr
Jean-Marie Launay PhD. 7, 8, 9, Jean-Etienne Bazin M.D., PhD. 1, Valerie Sautou PharmD.,
us
PhD. 3,4, Alexandre Mebazaa M.D., PhD. 5,7,8, Pascal Houze Pharm.D.10, Jean-Michel
CHU Clermont-Ferrand, Intensive Care Unit, Department of Peri-Operative
M
1
an
Constantin M.D., PhD. 1,2, Matthieu Legrand M.D., PhD. 5,7,8 for the PRONOBURN group
Univ. Clermont 1, D2R2, EA-7281, Faculty of Medicine, 28 Place Henri Dunant,
Ac ce pt e
2
d
Medecine, 63000 Clermont-Ferrand, France, and
F-63001 Clermont-Ferrand, France 3
CHU Clermont-Ferrand, Pôle Pharmacie, 58 Rue Montalembert, 63003 Clermont-
Ferrand, France 4
Clermont Université, Université d’Auvergne, EA4676C-BIOSENSS, 28 Place
Henri Dunant, 63000 Clermont-Ferrand, France 5
AP-HP, St-Louis Hospital, Department of Anesthesiology and Critical Care and
Burn Unit, 1 Avenue Claude-Vellefaux, 75010 Paris, France 6
CHU Clermont-Ferrand, Biostatistics Unit, DRCI, Gabriel-Montpied Hospital, 58
Rue Montalembert, F-63003 Clermont-Ferrand, France
Page 1 of 28
7
UMR INSERM 942, Institut National de la Santé et de la Recherche Médicale
(INSERM), Lariboisière hospital, 2 Rue Ambroise Paré, 75475 Paris Cedex 10 8
Univ Paris Diderot, Sorbonne Paris Cité, 1, Avenue Claude Vellefaux 75475
9
ip t
PARIS CEDEX 10, Paris, France AP-HP, Service de Biochimie, Lariboisière hospital, 2 Rue Ambroise Paré, 75475
AP-HP, St-Louis hospital, Department of Pharmacology, 1 Avenue Claude-
us
10
cr
Paris Cedex 10, France
Vellefaux, 75010 Paris, France
an
Corresponding author: Pr Matthieu Legrand, Department of Anesthesiology and Critical
M
Care and Burn Unit, St-Louis Hospital, Assistance Publique- Hopitaux de Paris; France. email address: [email protected], Tel: +33 (0)1 42 49 43 48, Fax : + 33 (0)1 42 39 99
Ac ce pt e
d
89
Financial disclosures: Equipment (filters) was provided by Sedana Medical AB, Uppsala, Sweden. This study was partially supported by a grant from “Fondation des gueules cassées”. The funders had no influence on study design, conduct, data analysis or in the preparation of this article.
Abstract
Page 2 of 28
Background: Sevoflurane has anti-inflammatory proprieties and short lasting effects making it of interest for procedural sedation in critically ill patients. We evaluated the pharmacokinetics of sevoflurane and metabolites in severely ill burn patients and controls. The secondary objective was to assess potential kidney injury.
ip t
Methods: prospective interventional study in a burn and a surgical intensive care unit 24 mechanically ventilated critically ill patients (12 burns, 12 controls) were included,. The
cr
sevoflurane was administered with an expired fraction target of 2% during short-term
us
procedural sedation. Plasma concentrations of sevoflurane, hexafluoroisopropanolol (HFIP) and free fluoride ions were recorded at different times. Kinetic Pro (Wgroupe, France) was
an
used for pharmacokinetic analysis. Kidney injury was assessed with neutrophil gelatinase associated lipocalin (NGAL).
M
Results: The mean total burn surface area was 36 ± 11%. The average plasma concentration of sevoflurane was 70.4 ± 37.5 mg.L-1 in burns and 57.2 ± 28.1 mg.L-1 in controls at the end
d
of the procedure (p = 0.58). The volume of distribution was higher (46.8 ± 7.2 vs 22.2 ± 2.50 l,
Ac ce pt e
p<0.001), and the drug half-life longer in burns (1.19 ± 0.28 hours vs 0.65 ± 0.04 h, p<0.0001). Free metabolite HFIP was higher in burns. Plasma fluoride was not different between burns and controls. NGAL didn’t rise after procedures. Conclusion: We observed an increased volume of distribution, slower elimination rate, and altered metabolism of sevoflurane in burn patients compared to controls. Repeated use for procedural sedation in burn patients needs further evaluation. No renal toxicity was detected. Trial registry number: ClinicalTrials.gov Identifier NCT02048683 Key words: sedation, intensive care, distribution volume, metabolism, burns The authors have no conflict of interest to declare related to this work. Of note, the Equipment (filters) was provided by Sedana Medical AB, Uppsala, Sweden. This study was partially supported by a grant from “Fondation des gueules cassées”. The funders had no influence on study design, conduct, data
Page 3 of 28
The study received approval of our Institutional Review Board in May 2013 (appropriate Institutional Review Board (Comité de Protection des Personnes Sud-Est VI, ClermontFerrand, France)) and from Agence Nationale de Sécurité du Médicament (ANSM) in June 2013 (No. EudraCT 2013-000960-26). The promoter was the University Hospital of (RBHP
2013
PERBET).
The
study
was
also
referenced
in
ip t
Clermont-Ferrand
ClinicalTrials.gov in January 2014 (identifier: NCT02048683). This manuscript adheres to the
cr
applicable EQUATOR guidelines.
us
Sevoflurane for procedural sedation in critically ill patients: a pharmacokinetic
an
comparative study between burn and non-burn patients.
Sebastien Perbet M.D. 1,2, Daniel Bourdeaux PharmD.,PhD. 3,4, Alexandre Lenoire M.D. 5,
M
Claire Biboulet M.D. 1, Bruno Pereira PhD. 6, Malha Sadoune 7, Benoit Plaud M.D., PhD. 5,7,8,
d
Jean-Marie Launay PhD. 7, 8, 9, Jean-Etienne Bazin M.D., PhD. 1, Valerie Sautou PharmD.,
Ac ce pt e
PhD. 3,4, Alexandre Mebazaa M.D., PhD. 5,7,8, Pascal Houze Pharm.D.10, Jean-Michel Constantin M.D., PhD. 1,2, Matthieu Legrand M.D., PhD. 5,7,8 for the PRONOBURN group
1
CHU Clermont-Ferrand, Intensive Care Unit, Department of Peri-Operative
Medicine, 63000 Clermont-Ferrand, France, and 2
Univ. Clermont 1, D2R2, EA-7281, Faculty of Medicine, 28 Place Henri Dunant,
F-63001 Clermont-Ferrand, France 3
CHU Clermont-Ferrand, Pharmacy department, 58 Rue Montalembert, 63003
Clermont-Ferrand, France
Page 4 of 28
4
Clermont University, Université d’Auvergne, EA4676C-BIOSENSS, 28 Place
Henri Dunant, 63000 Clermont-Ferrand, France 5
AP-HP, St-Louis Hospital, Department of Anaesthesiology and Critical Care and
6
ip t
Burn Unit, 1 Avenue Claude-Vellefaux, 75010 Paris, France CHU Clermont-Ferrand, Biostatistics Unit, DRCI, Gabriel-Montpied Hospital, 58
cr
Rue Montalembert, F-63003 Clermont-Ferrand, France 7
us
UMR INSERM 942, Institut National de la Santé et de la Recherche Médicale
(INSERM), Lariboisière hospital, 2 Rue Ambroise Paré, 75475 Paris Cedex 10 8
an
Univ Paris Diderot, Sorbonne Paris Cité, 1, Avenue Claude Vellefaux 75475
9
M
PARIS CEDEX 10, Paris, France
AP-HP, Department of biochemistry, Lariboisière hospital, 2 Rue Ambroise Paré,
AP-HP, St-Louis hospital, Department of Pharmacology, 1 Avenue Claude-
Ac ce pt e
10
d
75475 Paris Cedex 10, France
Vellefaux, 75010 Paris, France
Corresponding author: Pr Matthieu Legrand, Department of Anaesthesiology and Critical Care and Burn Unit, St-Louis Hospital, Assistance Publique- Hopitaux de Paris; France. email address: [email protected], Tel: +33 (0)1 42 49 43 48, Fax : + 33 (0)1 42 39 99 89
Financial disclosures: Equipment (filters) was provided by Sedana Medical AB, Uppsala, Sweden. This study was partially supported by a grant from “Fondation des gueules cassées”. The funders had no influence on study design, conduct, data analysis or in the preparation of this article.
Page 5 of 28
ip t cr
Abstract
Background: Sevoflurane has anti-inflammatory proprieties and short lasting effects making
us
it of interest for procedural sedation in critically ill patients. We evaluated the pharmacokinetics of sevoflurane and metabolites in severely ill burn patients and controls.
an
The secondary objective was to assess potential kidney injury.
M
Methods: prospective interventional study in a burn and a surgical intensive care unit; 24 mechanically ventilated critically ill patients (12 burns, 12 controls) were included. The
d
sevoflurane was administered with an expired fraction target of 2% during short-term
Ac ce pt e
procedural sedation. Plasma concentrations of sevoflurane, hexafluoroisopropanolol (HFIP) and free fluoride ions were recorded at different times. Kinetic Pro (Wgroupe, France) was used for pharmacokinetic analysis. Kidney injury was assessed with neutrophil gelatinase associated lipocalin (NGAL).
Results: The mean total burn surface area was 36 ± 11%. The average plasma concentration of sevoflurane was 70.4 ± 37.5 mg.L-1 in burns and 57.2 ± 28.1 mg.L-1 in controls at the end of the procedure (p = 0.58). The volume of distribution was higher (46.8 ± 7.2 vs 22.2 ± 2.50 l, p<0.001), and the drug half-life longer in burns (1.19 ± 0.28 hours vs 0.65 ± 0.04 h, p<0.0001). Free metabolite HFIP was higher in burns. Plasma fluoride was not different between burns and controls. NGAL didn’t rise after procedures.
Page 6 of 28
Conclusion: We observed an increased volume of distribution, slower elimination rate, and altered metabolism of sevoflurane in burn patients compared to controls. Repeated use for procedural sedation in burn patients needs further evaluation. No renal toxicity was detected. Trial registry number: ClinicalTrials.gov Identifier NCT02048683
Ac ce pt e
d
M
an
us
cr
ip t
Key words: sedation, intensive care, distribution volume, metabolism, burns
Page 7 of 28
Introduction Due to restriction in the use of continuous sedation in the critically ills, procedural sedation protocols for invasive procedures might be frequently indicated[1][2]. Invasive or painful procedures in the intensive care unit (ICU) may require sedation to improve patient comfort
ip t
and safety. However, the potential benefits of sedation should be balanced by potential risks,
cr
including prolonged sedation and the need for prolonged invasive mechanical ventilation. Burn patients require frequent dressing changes, which may require deep sedation[3]. The use
us
of sedative drugs with short half-life allows the patient to awake rapidly after the procedure and avoids unnecessary prolonged deep sedation[1]. Sevoflurane is a halogen gas with anti-
an
inflammatory proprieties and a short half-life and low accumulation in fat tissue, which makes
M
it a potentially appropriate sedative agent in this setting[4][5]. Its elimination produces metabolites as fluoride and hexafluoroisopropanolol (HFIP) with potential activity and
d
toxicity.
Ac ce pt e
The use of halogenated gas is facilitated via the ICU Anaconda® (or Anaesthetic Conserving Device (ACD), Sedana Medical, Sundbyberg, Sweden) or Mirus® (Pall Medical, Dreieich, Germany)[6]. The Anaconda® is a bulky medical device inserted between the patient and the ventilator that enables the administration and recycling of gas. This device also humidifies and warms up the inspired gases.
The pharmacokinetics of intravenous drugs can be modified in burn patients due to changes in the distribution volume and haemodynamic parameters[7][8]. However, there are no data available regarding the pharmacokinetic changes for inhaled sedative agents in these patients. Thus, the primary objective of this study was to evaluate the pharmacokinetics of sevoflurane for procedural sedation in severely ill burn patients and to compare the results with non-burn critically ill patients in the ICU. The secondary objective was to identify potential renal toxicity.
Page 8 of 28
Methods Patients The following study inclusion criteria were used: burn patients over 18 years of age, total burn surface area (TBSA) between 20 and 50%, and patients requiring mechanical ventilation with sedation for a predictable period of 2 to 4 hours in burn patients within 15 days of burn
ip t
injury. The controls were patients requiring mechanical ventilation with procedural sedation
cr
control for an expected period of 2 to 4 hours. The exclusion criteria were pregnancy, allergy to halogenated anaesthetic agents, inclusion in other protocols, personal or family history of
us
malignant hyperthermia, severe liver failure (factor V <50% or alanine transaminase> 500 IU. L-1), patients deprived of liberty or under guardianship, medication with isoniazid treatment,
an
diethyldithiocarbamate or disulfiram, or if the patient or next of kin refused to participate.
M
The patients (or next of kin for patients unable to consent) provided written informed consent. The burn patients were from the burn intensive care unit of St-Louis hospital, Paris, France.
d
The control ICU patients were included from the surgical ICU of Clermont-Ferrand hospital.
Ac ce pt e
The study received approval of our Institutional Review Board in May 2013 (appropriate Institutional Review Board (Comité de Protection des Personnes Sud-Est VI, ClermontFerrand, France)) and from the Agence Nationale de Sécurité du Médicament (ANSM) in June 2013 (No. EudraCT 2013-000960-26). The promoter was the University Hospital of Clermont-Ferrand
(RBHP
2013
PERBET).
The
study
was
also
referenced
in
ClinicalTrials.gov in January 2014 (identifier: NCT02048683). This manuscript adheres to the applicable EQUATOR guidelines.
The control patients were matched to burn patients for gender, age (+/- 5 years) and body mass index (BMI) (+/- 2.5 kg. m-2). Protocol All patients were managed using volume-controlled ventilation with protective ventilation. Both the ventilator frequency and tidal volume (6-8 mL.kg-1 of ideal body weight) remained
Page 9 of 28
unchanged during the sevoflurane wash-in and washout periods. The patients received continuous intravenous remifentanil infusion for painful procedures to obtain a behavioural pain score less than 3. The maintenance of sedation was achieved by inhalation of sevoflurane via the Anaconda® system with a fraction of 2% expired. We increased the flow of the
ip t
syringe pump from 6 to 10 ml.h-1 to progressively increase the fraction of sevoflurane from 0 to 2%. We collected blood samples at baseline and then after 5 min, 15 min, and 1 hour after
cr
detection (by the monitor) of sevoflurane (0.1%) in the breathing circuit. We also collected
us
blood samples immediately prior to stopping inhalation and then at 5, 10, 30, 60, and 120 minutes after cessation of sevoflurane administration. Additional samples were collected at 5
an
min, 10 min, 30 min, 2 h, 4 h, and 6 h after cessation of sevoflurane delivery (Figure 1). The blood was collected via a venous catheter directly into a 5-ml vacuum tube. We were careful
M
to limit residual air as previously described. Each sample was promptly centrifuged for 3 min at 5,000 rpm. We transferred a 1-ml sample of plasma to a 20-ml headspace tube (Antélia,
d
Décines-Charpieu, France) and then added 20 μl of an internal chromatographic standard
Ac ce pt e
solution (600 µg/ml chloroform in undecane). The tube was immediately hermetically sealed using a Teflon-sealed cap and frozen at -20°C until analysis the following week. All the samples were collected after the first sevoflurane administration. We noted the end-tidal CO2 and end-tidal sevoflurane concentrations at the time of each blood sample collection. The expiration volume/min was measured by the ventilator and was also recorded. The tubing running from the syringe to the ACD was empty at the commencement of infusion to ensure the complete absence of sevoflurane until the required time of delivery. The time of first detection of gas served as the start of kinetic calculations. The urinary and serum neutrophil gelatinase-associated lipocalin (NGAL) were collected before sedation and again at 24-h and 48-h after sedation. The serum creatinine values were monitored for 7 days following inclusion. Acute kidney injury (AKI) was defined according
Page 10 of 28
to the Kidney Disease Global Outcome, KDIGO criteria11. The baseline serum creatinine value was the value on the day of inclusion. Plasma Sevoflurane, HFIP, and fluoride quantitation Sevoflurane and HFIP were quantified using a gas chromatograph (Autosystem, Perkin Elmer Corp., San Jose, US) coupled to a
ip t
headspace system (HS40, Perkin Elmer Corp., San Jose, US) and a mass spectrometer (Clarus 500, Perkin Elmer Corp., San Jose, US) as previously described[9][6]. The mean values for
cr
accuracy and precision results obtained from the method validation process for sevoflurane
us
were 2.2% and 6.9%, respectively. The values for HFIP were 5.8% and 7.2%[10]. The limit of quantification was determined at 1 µg.mL-1 for the two molecules.
an
The total plasma fluoride levels were measured at 0 h, 24 h, and 48 h after the initiation of sevoflurane sedation. An ionometric detector (Sevenmulti, Mettler Toledo, Viroflay, France)
M
fitted with a fluoride specific electrode (Mettler Toledo, Viroflay, France) was used to quantify plasma fluoride levels. We then constructed a calibration curve using fluoride
d
standards ranging from 1-70 µmol.L-1. Each 4-ml volume of patient plasma was treated with
Ac ce pt e
400 microliters of TISAB activation solution (Mettler Toledo, Viroflay, France) prior to analysis. The precision and accuracy of the method were verified. The mean values for accuracy and precision results for fluoride obtained from the method validation process were 2% and 4.5%, respectively. Pharmacokinetic model
The aim of the present study was to establish a pharmacokinetic model of sevoflurane used with the ACD for sedation in severely ill burn patients without hepatic or renal failure. We then compared the pharmacokinetic parameters with non-burn critically ill patients. The pharmacokinetic analysis was performed with KineticPro software (Wgroupe, Pommiers la Placette, France)[6][11][12]. The observed profile (time versus observed concentrations curve) was fitted by compartmental models involving one to three compartments. As
Page 11 of 28
previously observed, the best fit was obtained by the 2-compartment model (the 3compartment was over-parameterized, and the 1-compartment model was mis-specified)[13]. The terms y1 and y2 represent the concentrations in the central and peripheral compartments, respectively. Vdis is the volume of distribution of the central compartment, and ke is the
ip t
elimination rate. The terms k1-2 and k2-1 show the exchange rates with the peripheral compartment. These parameters were estimated by nonlinear regression of y1 to the observed
cr
profile. The log-linear regression of the terminal portion of the blood concentration-time
us
curve was used to estimate the half-life of sevoflurane (T1/2). The total clearance (Cl=ke. Vdis) was also calculated.
an
We recorded sevoflurane syringe flow over time and calculated the total amount of sevoflurane administered to evaluate the amount of sevoflurane effectively absorbed. A 10%
M
leakage of sevoflurane via the ACD was effectively assumed[9]. The samples were collected
Statistical analysis
d
until 6 hours after cessation of sevoflurane delivery.
Ac ce pt e
The sample size estimation was difficult because both the between and within patient variability were unknown. However, we conducted (a priori) simulations to determine whether the sample size was adequate to allow changes to be detected over time and to estimate an intraclass correlation coefficient (ICC). We considered the results obtained for normal weight patients with standard half-life (2.84 with a standard deviation of 0.37) to select a sample size of 12 patients in each group. For 9 time points, such sample size allow to highlight a minimal difference between groups equals 0.43 with a two-sided type I error at 5%, a statistical power greater than 80% and an ICC greater than 0.8 [14].
The statistical analysis was performed using Stata software (version 13, StataCorp, College Station, US). The tests were two-sided, and the type I error was set at α=0.05. The baseline
Page 12 of 28
characteristics are presented as the mean (standard deviation (SD)) or the median [interquartile range] for continuous data (assumption of normality assessed by using the Shapiro-Wilk test). The data are shown as the number of patients and associated percentages for categorical parameters. All quantitative data were compared between groups (Burns vs. Controls) using
ip t
Student’s t-test or Mann-Whitney test when the t-test assumptions were not satisfied. The assumption of homoscedasticity was studied using the Fisher-Snedecor’s test. The
cr
comparisons between groups for categorical data were performed using either Chi-squared or
us
Fisher’s exact tests. We accounted for the between and within subject variability (due to several repeated measures for a same patient) using random-effects models (linear for
an
quantitative dependant outcome, with logarithmic transformation to achieve normality when appropriate) rather than usual statistical tests that are not appropriate because the hypothesis
M
of independent data was not verified. The patients were studied as random effects, and then
Ac ce pt e
d
the following fixed effects were analysed: group, time point evaluation, and interactions.
Page 13 of 28
Results Patients This study enrolled 24 patients (12 burns, 12 controls) between December 2013 and August 2015. In the burn patients, the TBSA was 36 ±11 %. The reasons for ICU admission in the
ip t
control subjects were related to surgery in five patients and were medical in seven patients.
cr
The control and burn patients were not different regarding gender, age, and BMI (Table 1). The controls had non-significantly higher Simplified Acute Physiology Score II (SAPSII) (45
us
± 12 versus 35 ± 20), p = 0.27), higher serum creatinine, and impaired renal function (92 ± 30 μmol.L-1 vs. 71 ± 39 μmol.L-1, p = 0.18) compared to the burn cases. One patient in each
an
group died while in the ICU.
M
Pharmacokinetics of sevoflurane
The plateau plasma concentrations of sevoflurane were not significantly different between
d
groups. However, the drug concentrations were higher in burn cases during the washout phase.
Ac ce pt e
The concentration increased rapidly to 73 ± 38 mg.L-1 and 57 ± 28 mg.L-1 in burn cases and controls at the end of the administration of sevoflurane (p=0.27) (Figure 2), respectively. The rapid plasma sevoflurane washout was associated with a rapid decrease of sevoflurane concentration after the end of sevoflurane inhalation (Figure 2). During the procedure, the subjects inhaled 61.01 ± 26.62 mg of sevoflurane in 181 ± 62 minutes and 59.33 ± 19.86 mg sevoflurane in 182 ± 65 minutes in burn cases and controls, respectively (p = 0.51). The expired fractions of sevoflurane were similar between groups. Table 2 describes the 2-compartment pharmacokinetic model constants calculated by the pharmacokinetic software. The burn cases had higher Vdis, lower ke, longer k1-2 and k2-1, longer T1/2 and lower Cl.
Page 14 of 28
The concentration of sevoflurane was correlated with the expired fraction of sevoflurane both in burn cases and controls (p < 0.01). The evolution of ventilation parameters is described in supplementary file 1. Plasma hexafluoroisopropanolol and fluoride
ip t
The level of free HFIP was significantly higher in burn cases than controls (p=0.02, Figure 3). The maximum values of plasma concentration of free HFIP were 5.05 ± 2.26 mg.mL-1 at 1
cr
hour and 1.63 ± 1.19 mg.mL-1 at 30 minutes after discontinuation of sedation sevoflurane in
us
burn cases and controls, respectively. The mean plasma fluoride values peaked 24 h after the end of inhalation, with a maximum of 22 ± 14 mg.mL-1 in burn cases and 38 ± 34 mg.mL-1 in
an
controls (p=0.16). Renal function and injury
M
The controls had higher serum creatinine levels than the burn cases throughout the 7-day observational period following inclusion. However, there were no significant changes
d
observed within groups (Supplementary file 2). The renal function was not significantly
Ac ce pt e
altered in the days following sedation with sevoflurane in burn cases. One burn patient (8%) and four (33%) controls met the criteria for AKI during the 7 days following inclusion. The cause of AKI in the burn patient was attributed to septic shock. Interestingly, we did not observe significant elevation of plasma and urine NGAL after inclusion in both groups. The urine and plasma NGAL remained < 150 ng.mL-1 in all patients except one (maximum 154 ng.mL-1) (Figure 4).
Page 15 of 28
Discussion In this study, we compared the pharmacokinetics of inhaled sevoflurane for short-term procedural sedation in severely ill burn patients and non-burn critically ill patients. We observed a rapid rise in sevoflurane plasma concentration after initiation of sedation in both
ip t
groups. However, burn patients showed higher volume of distribution and lower clearance.
cr
Furthermore, the higher HFIP levels suggest there is altered hepatic metabolism in burn
the stable plasma and urine NGAL level after procedures.
us
patients. Finally, no evidence of kidney injury could be observed in this cohort as shown by
Prolonged deep sedation is no longer recommended in critically ill patients[15]. However,
an
short-term deep sedation may be used to improve patient comfort and safety in painful or
M
invasive procedures. Burn patients require iterative procedural sedation for dressing changes. The inhaled sevoflurane appears to have appropriate pharmacokinetics and allows for short-
d
term sedation with theoretical rapid clearance[6]. The pharmacokinetics of several
Ac ce pt e
intravenous medications is altered in burn patients with increased distribution volume but no pharmacokinetics data are available regarding inhaled sedation with sevoflurane. Despite the pharmacokinetic properties that make inhaled sevoflurane suitable for short-term sedation (rapid increase of plasma concentration and rapid plasma clearance), there were significant differences compared to non-burn critically ill patients matched for age, gender, and BMI. The two important features were increased plasma sevoflurane concentration at the end of the procedure, despite similar inhaled doses and duration of sedation. The reasons for the higher concentration might be related to higher cardiac output and lung perfusion increasing the contact between alveolar gases and blood in burn patients compared to controls and primary injury to the lungs and airways from direct thermal inhalation, or widespread systemic inflammation[16]. Second, clearance of sevoflurane was lower with higher half-life in burn cases, which could be associated with a renal or non-renal clearance modification.
Page 16 of 28
However, there was a rapid decrease in plasma concentration observed. This decrease was associated with a significant increase in distribution volume generally observed in burn patients with increased capillary permeability and fluid extravasation[17]. These explanations are speculative because we could not determine the blood/gas coefficients for sevoflurane in
ip t
the burn and control patients. The concentration of free HFIP was higher in the burn patients. The reason for this
cr
observation could be modified by hypoalbuminaemia related to the loss of protein-rich fluid
us
via the burn wound, a decrease in constitutive hepatic protein synthesis, and an increase in catabolism (with an increase of the unbound (free) fraction of a drug) and by hepatic
an
metabolism of sevoflurane with lower glucuroconjugation and/or overall decrease in renal elimination[18]. The possibility of altered renal clearance is unlikely because the renal
M
function was normal in the burn patients in this cohort. Conversely, hepatic metabolism has previously been found to be impaired after burn injury[19][20][18]. Further studies should
d
explore sevoflurane hepatic metabolism in burn cases for longer periods and/or following
Ac ce pt e
iterative sedation with sevoflurane and exclude the potential accumulation of its metabolites. A decreased clearance may lead to accumulation of sevoflurane during prolonged and/or iterative administration in burn patients and raises concerns regarding safety in this specific population.
In our study, there was no significant impact of sevoflurane on kidney injury. Although several patients did develop AKI, ICU patients have many factors associated with a risk of AKI (e.g., sepsis, trauma, hemodynamic, nephrotoxic, etc.). In this study, we measured plasma and urine NGAL. NGAL is a highly sensitive biomarker of AKI after sevoflurane administration[21]. NGAL did not significantly rise within 48 hours after sevoflurane administration. We used a cut-off of 150 ng.mL-1, which has been reported as highly sensitive to AKI,[22][23] and found all patients except one (with a peak at 154 ng.mL-1) did
Page 17 of 28
not reach this cut-off. Therefore, it is unlikely that sevoflurane promoted the development of AKI in these cohorts. Finally, the fluoride values remained low. In contrast to the renal metabolism of the methoxyflurane that may have resulted in renal intraparenchymal concentrations above 50
ip t
µmol.L-1 [24], the hepatic metabolism of the sevoflurane may explain the renal toxicity of
cr
fluoride when high fluoride levels are not observed. Renal toxicity was however not observed in our study.
us
Sevoflurane therefore appears to have slight different metabolism in burn and non-burn patients but the pharmacokinetics however appear to be safe – especially with no detection of
an
kidney injury – using transient procedural sedation in both populations. Based on this results,
M
we feel that sevoflurane could be used safely in both population in this setting for procedural sedation but that safety of continuous use or multiple procedural sedation with <48h between
Ac ce pt e
d
administration would require additional investigations.
Our study has several limitations. Our data shows a wide range of values for the plasma sevoflurane concentration and its metabolite, which suggests there is inter-individual variability of sevoflurane pharmacokinetics. The intra-individual variability was not explored in this study. Although this was a multicentre study, all burn cases and controls were included in a single centre. Finally, only short-term sedation with samples until only 6 hours after cessation of sevoflurane delivery were studied, and the results could not be extrapolated to long-term sedation or iterative procedures.
To conclude, we confirm rapid clearance of sevoflurane in non-burn critically ill patients during procedural sedation. We however observed altered metabolism and accumulation of HFIP, larger volume of distribution, and lower clearances of sevoflurane in burn patients.
Page 18 of 28
These differences must be considered while using sevoflurane for sedation in burn patients. Impact of iterative procedural sedations on sevoflurane and its metabolites clearance should
Ac ce pt e
d
M
an
us
cr
ip t
be explored.
Page 19 of 28
Acknowledgements: We thank the nurses, residents, and physicians who provided care for the study patients and collected data. We also thank Dr Hélène Sauvageon (pharmacy, St-Louis Hospital, Paris,
Ac ce pt e
d
M
an
us
cr
ip t
France) for fluoride dosages.
Page 20 of 28
References
Ac ce pt e
d
M
an
us
cr
ip t
[1] Barr J, Fraser GL, Puntillo K, Ely EW, Gélinas C, Dasta JF, et al. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med 2013;41:263–306. doi:10.1097/CCM.0b013e3182783b72. [2] Hughes CG, Girard TD, Pandharipande PP. Daily sedation interruption versus targeted light sedation strategies in ICU patients. Crit Care Med 2013;41:S39–45. doi:10.1097/CCM.0b013e3182a168c5. [3] Gregoretti C, Decaroli D, Piacevoli Q, Mistretta A, Barzaghi N, Luxardo N, et al. Analgo-sedation of patients with burns outside the operating room. Drugs 2008;68:2427–43. [4] Jabaudon M, Boucher P, Imhoff E, Chabanne R, Faure J-S, Roszyk L, et al. Sevoflurane for Sedation in Acute Respiratory Distress Syndrome. A Randomized Controlled Pilot Study. Am J Respir Crit Care Med 2017;195:792–800. doi:10.1164/rccm.2016040686OC. [5] Chabanne R, Perbet S, Futier E, Ben Said NA, Jaber S, Bazin J-E, et al. Impact of the anesthetic conserving device on respiratory parameters and work of breathing in critically ill patients under light sedation with sevoflurane. Anesthesiology 2014;121:808–16. doi:10.1097/ALN.0000000000000394. [6] Perbet S, Bourdeaux D, Sautou V, Pereira B, Chabanne R, Constantin JM, et al. A pharmacokinetic study of 48-hour sevoflurane inhalation using a disposable delivery system (AnaConDa®) in ICU patients. Minerva Anestesiol 2014;80:655–65. [7] Jaehde U, Sörgel F. Clinical pharmacokinetics in patients with burns. Clin Pharmacokinet 1995;29:15–28. doi:10.2165/00003088-199529010-00003. [8] Mesnil M, Capdevila X, Bringuier S, Trine P-O, Falquet Y, Charbit J, et al. Long-term sedation in intensive care unit: a randomized comparison between inhaled sevoflurane and intravenous propofol or midazolam. Intensive Care Med 2011;37:933–41. doi:10.1007/s00134-011-2187-3. [9] Meiser A, Bellgardt M, Belda J, Röhm K, Laubenthal H, Sirtl C. Technical performance and reflection capacity of the anaesthetic conserving device--a bench study with isoflurane and sevoflurane. J Clin Monit Comput 2009;23:11–9. doi:10.1007/s10877-0089158-4. [10] Bourdeaux D, Sautou-Miranda V, Montagner A, Perbet S, Constantin JM, Bazin J-E, et al. Simple assay of plasma sevoflurane and its metabolite hexafluoroisopropanol by headspace GC-MS. J Chromatogr B Analyt Technol Biomed Life Sci 2010;878:45–50. doi:10.1016/j.jchromb.2009.11.018. [11] Iliadis A, Brown AC, Huggins ML. APIS: a software for model identification, simulation and dosage regimen calculations in clinical and experimental pharmacokinetics. Comput Methods Programs Biomed 1992;38:227–39. [12] Petricoul O, Claret L, Barbolosi D, Iliadis A, Puozzo C. Information tools for exploratory data analysis in population pharmacokinetics. J Pharmacokinet Pharmacodyn 2001;28:577–99. [13] Enlund M, Kietzmann D, Bouillon T, Züchner K, Meineke I. Population pharmacokinetics of sevoflurane in conjunction with the AnaConDa: toward target-controlled infusion of volatiles into the breathing system. Acta Anaesthesiol Scand 2008;52:553–60. doi:10.1111/j.1399-6576.2008.01579.x. [14] Ogungbenro K, Aarons L. How many subjects are necessary for population pharmacokinetic experiments? Confidence interval approach. Eur J Clin Pharmacol 2008;64:705–13. doi:10.1007/s00228-008-0493-7.
Page 21 of 28
Ac ce pt e
d
M
an
us
cr
ip t
[15] Barr J, Pandharipande PP. The pain, agitation, and delirium care bundle: synergistic benefits of implementing the 2013 Pain, Agitation, and Delirium Guidelines in an integrated and interdisciplinary fashion. Crit Care Med 2013;41:S99–115. doi:10.1097/CCM.0b013e3182a16ff0. [16] Enkhbaatar P, Traber DL. Pathophysiology of acute lung injury in combined burn and smoke inhalation injury. Clin Sci Lond Engl 1979 2004;107:137–43. doi:10.1042/CS20040135. [17] Udy AA, Roberts JA, Lipman J, Blot S. The effects of major burn related pathophysiological changes on the pharmacokinetics and pharmacodynamics of drug use: An appraisal utilizing antibiotics. Adv Drug Deliv Rev 2018;123:65–74. doi:10.1016/j.addr.2017.09.019. [18] Yogaratnam D, Ditch K, Medeiros K, Miller MA, Smith BS. The Impact of Liver and Renal Dysfunction on the Pharmacokinetics and Pharmacodynamics of Sedative and Analgesic Drugs in Critically Ill Adult Patients. Crit Care Nurs Clin North Am 2016;28:183– 94. doi:10.1016/j.cnc.2016.02.009. [19] Pereira CT, Herndon DN. The pharmacologic modulation of the hypermetabolic response to burns. Adv Surg 2005;39:245–61. [20] Finnerty CC, Jeschke MG, Qian W-J, Kaushal A, Xiao W, Liu T, et al. Determination of burn patient outcome by large-scale quantitative discovery proteomics. Crit Care Med 2013;41:1421–34. doi:10.1097/CCM.0b013e31827c072e. [21] Ronco C, Legrand M, Goldstein SL, Hur M, Tran N, Howell EC, et al. Neutrophil gelatinase-associated lipocalin: ready for routine clinical use? An international perspective. Blood Purif 2014;37:271–85. doi:10.1159/000360689. [22] Zarbock A, Kellum JA, Schmidt C, Van Aken H, Wempe C, Pavenstädt H, et al. Effect of Early vs Delayed Initiation of Renal Replacement Therapy on Mortality in Critically Ill Patients With Acute Kidney Injury: The ELAIN Randomized Clinical Trial. JAMA 2016;315:2190–9. doi:10.1001/jama.2016.5828. [23] Legrand M, Mebazaa A, Ronco C, Januzzi JL. When cardiac failure, kidney dysfunction, and kidney injury intersect in acute conditions: the case of cardiorenal syndrome. Crit Care Med 2014;42:2109–17. doi:10.1097/CCM.0000000000000404. [24] Kharasch ED, Hankins DC, Thummel KE. Human kidney methoxyflurane and sevoflurane metabolism. Intrarenal fluoride production as a possible mechanism of methoxyflurane nephrotoxicity. Anesthesiology 1995;82:689–99.
Page 22 of 28
Tables Table 1. Patient characteristics. BMI, body mass index; TBSA, total body surface area; SAPS, simplified acute physiology score; GFR, glomerular filtration rate. Data expressed in mean (SD), (SD, standard deviation). Controls (n=12) 10 (83%) 55 (17) 71,75 (11,87) 1,70 (0,08) 24,66 (3,31) NA 42,83 (13,78) 10,95 (3,52) 92,18 (30,28) 81,71 (29,1) 182,25 (64,92)
an
us
cr
ip t
Burns (n= 12) 8 (67%) 49 (17) 80,25 (12,44) 1,74 (0,12) 26,27 (3,07) 36,13 (11,11) 34,50 (19,84) 7,84 (4,80) 71,33 (39,88) 133,83 (76,48) 181,33 (62,08)
Male sex - n (%) Age (years) Weight (kg)Height (m) - mean (SD) BMI (kg.m-2) TBSA (%) SAPS 2 Plasma urea (mmol.L-1) at inclusion Plasma creatinine (µmol.L-1) at inclusion GFR (ml.min-1) Sedation duration (min)
Ke (h-1)
46.80 (7.20) 22.20 (2.50) <0.0001
5.95 (0.92) 14.70 (1.61) <0.0001
K1-2 (h-1)
K2-1 (h-1)
T½ (h)
Clpl (l/h)
RE
2.44 (0.73) 4.80 (0.94) <0.0001
0.85 (0.24) 1.44 (0.12) <0.0001
1.19 (0.28) 0.65 (0.04) <0.0001
279.00 (61.10) 327.00 (51.90) 0.04
13.98% 7.55%
Ac ce pt e
Burns (n=12) Controls (n=12) P
Vdis (l)
d
M
Table 2. Pharmacokinetics of sevoflurane in burn cases and controls. The terms; y1 and y2 are the concentrations in the central and peripheral compartments, respectively; Vdis, the volume of distribution of the central compartment (in liters); ke, the elimination rate; k1-2 and k2-1, the exchange rates with the peripheral compartment; T1/2, the half-life of sevoflurane; Clpl, total plasma clearance; RE, residual error. Data expressed in mean (SD), (SD, standard deviation).
Page 23 of 28
Figures legends Figure 1: Study protocol (T0 = start of the sedation, T4 = end of sedation). Figure 2: Expired fraction (Exp, %) and plasma concentrations (mg.L-1) of sevoflurane in control and burn patients.
ip t
Figure 3: Plasma hexafluoroisopropanolol (HFIP) concentration (mg.L-1) in control and burn patients.
cr
Figure 4: Urine and plasma NGAL values (ng. L-1) in control and burn patients (values
Ac ce pt e
d
M
an
us
expressed in median and interquartilerange).
Page 24 of 28
Glossary
AKI, Acute kidney injury ACD, Anaesthetic Conserving Device
ip t
BMI, body mass index
cr
HFIP, hexafluoroisopropanolol NGAL, neutrophil gelatinase associated lipocalin
us
SAPSII, Simplified Acute Physiology Score II
an
TBSA, Total body surface area
d
M
Table 1. Patient characteristics. BMI, body mass index; TBSA, total body surface area; SAPS, simplified acute physiology score; GFR, glomerular filtration rate. Data expressed in mean (SD), (SD, standard deviation).
Ac ce pt e
Male sex - n (%) Age (years) Weight (kg)Height (m) - mean (SD) BMI (kg.m-2) TBSA (%) SAPS 2 Plasma urea (mmol.L-1) at inclusion Plasma creatinine (µmol.L-1) at inclusion GFR (ml.min-1) Sedation duration (min)
Burns (n= 12) 8 (67%) 49 (17) 80,25 (12,44) 1,74 (0,12) 26,27 (3,07) 36,13 (11,11) 34,50 (19,84) 7,84 (4,80) 71,33 (39,88) 133,83 (76,48) 181,33 (62,08)
Controls (n=12) 10 (83%) 55 (17) 71,75 (11,87) 1,70 (0,08) 24,66 (3,31) NA 42,83 (13,78) 10,95 (3,52) 92,18 (30,28) 81,71 (29,1) 182,25 (64,92)
Table 2. Pharmacokinetics of sevoflurane in burn cases and controls. The terms; y1 and y2 are the concentrations in the central and peripheral compartments, respectively; Vdis, the volume of distribution of the central compartment (in liters); ke, the elimination rate; k1-2 and k2-1, the exchange rates with the peripheral compartment; T1/2, the half-life of sevoflurane; Clpl, total plasma clearance; RE, residual error. Data expressed in mean (SD), (SD, standard deviation).
Vdis (l)
Ke (h-1)
K1-2 (h-1)
K2-1 (h-1)
T½ (h)
Clpl (l/h)
RE
Page 25 of 28
2.44 (0.73) 0.85 (0.24) 1.19 (0.28) 4.80 (0.94) 1.44 (0.12) 0.65 (0.04) <0.0001 <0.0001 <0.0001
279.00 (61.10) 327.00 (51.90) 0.04
13.98% 7.55%
d
M
an
us
cr
ip t
46.80 (7.20) 5.95 (0.92) 22.20 (2.50) 14.70 (1.61) <0.0001 <0.0001
Ac ce pt e
Burns (n=12) Controls (n=12) P
Page 26 of 28
Page 27 of 28
d
Ac ce pt e us
an
M
cr
ip t
2
1
0
D
D
0
0
2
urine NGAL (ng.mL-1)
ed controls burns
D
D
60
pt
ce
Ac
20
1
0
40
D
D
plasma NGAL (ng.mL-1) 80 100
80
controls burns
60
40
20
Page 28 of 28
S2352-5568(17)30244-8 https://doi.org/doi:10.1016/j.accpm.2018.02.001 ACCPM 344
To appear in: Received date: Revised date: Accepted date:
2-10-2017 6-2-2018 7-2-2018
Please cite this article as: Sebastien PerbetDaniel BourdeauxAlexandre LenoireClaire BibouletBruno PereiraMalha SadouneBenoit PlaudJean-Marie LaunayJean-Etienne BazinValerie SautouAlexandre MebazaaPascal HouzeJean-Michel ConstantinMatthieu Legrandfor the PRONOBURN group, Sevoflurane For Procedural Sedation In Critically Ill Patients: A Pharmacokinetic Comparative Study Between Burn And Non-Burn Patients (2018), https://doi.org/10.1016/j.accpm.2018.02.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
SEVOFLURANE FOR PROCEDURAL SEDATION IN CRITICALLY ILL PATIENTS: A PHARMACOKINETIC COMPARATIVE STUDY BETWEEN BURN AND NON-BURN PATIENTS
ip t
Sebastien Perbet M.D. 1,2, Daniel Bourdeaux PharmD.,PhD. 3,4, Alexandre Lenoire M.D. 5, Claire Biboulet M.D. 1, Bruno Pereira PhD. 6, Malha Sadoune 7, Benoit Plaud M.D., PhD. 5,7,8,
cr
Jean-Marie Launay PhD. 7, 8, 9, Jean-Etienne Bazin M.D., PhD. 1, Valerie Sautou PharmD.,
us
PhD. 3,4, Alexandre Mebazaa M.D., PhD. 5,7,8, Pascal Houze Pharm.D.10, Jean-Michel
CHU Clermont-Ferrand, Intensive Care Unit, Department of Peri-Operative
M
1
an
Constantin M.D., PhD. 1,2, Matthieu Legrand M.D., PhD. 5,7,8 for the PRONOBURN group
Univ. Clermont 1, D2R2, EA-7281, Faculty of Medicine, 28 Place Henri Dunant,
Ac ce pt e
2
d
Medecine, 63000 Clermont-Ferrand, France, and
F-63001 Clermont-Ferrand, France 3
CHU Clermont-Ferrand, Pôle Pharmacie, 58 Rue Montalembert, 63003 Clermont-
Ferrand, France 4
Clermont Université, Université d’Auvergne, EA4676C-BIOSENSS, 28 Place
Henri Dunant, 63000 Clermont-Ferrand, France 5
AP-HP, St-Louis Hospital, Department of Anesthesiology and Critical Care and
Burn Unit, 1 Avenue Claude-Vellefaux, 75010 Paris, France 6
CHU Clermont-Ferrand, Biostatistics Unit, DRCI, Gabriel-Montpied Hospital, 58
Rue Montalembert, F-63003 Clermont-Ferrand, France
Page 1 of 28
7
UMR INSERM 942, Institut National de la Santé et de la Recherche Médicale
(INSERM), Lariboisière hospital, 2 Rue Ambroise Paré, 75475 Paris Cedex 10 8
Univ Paris Diderot, Sorbonne Paris Cité, 1, Avenue Claude Vellefaux 75475
9
ip t
PARIS CEDEX 10, Paris, France AP-HP, Service de Biochimie, Lariboisière hospital, 2 Rue Ambroise Paré, 75475
AP-HP, St-Louis hospital, Department of Pharmacology, 1 Avenue Claude-
us
10
cr
Paris Cedex 10, France
Vellefaux, 75010 Paris, France
an
Corresponding author: Pr Matthieu Legrand, Department of Anesthesiology and Critical
M
Care and Burn Unit, St-Louis Hospital, Assistance Publique- Hopitaux de Paris; France. email address: [email protected], Tel: +33 (0)1 42 49 43 48, Fax : + 33 (0)1 42 39 99
Ac ce pt e
d
89
Financial disclosures: Equipment (filters) was provided by Sedana Medical AB, Uppsala, Sweden. This study was partially supported by a grant from “Fondation des gueules cassées”. The funders had no influence on study design, conduct, data analysis or in the preparation of this article.
Abstract
Page 2 of 28
Background: Sevoflurane has anti-inflammatory proprieties and short lasting effects making it of interest for procedural sedation in critically ill patients. We evaluated the pharmacokinetics of sevoflurane and metabolites in severely ill burn patients and controls. The secondary objective was to assess potential kidney injury.
ip t
Methods: prospective interventional study in a burn and a surgical intensive care unit 24 mechanically ventilated critically ill patients (12 burns, 12 controls) were included,. The
cr
sevoflurane was administered with an expired fraction target of 2% during short-term
us
procedural sedation. Plasma concentrations of sevoflurane, hexafluoroisopropanolol (HFIP) and free fluoride ions were recorded at different times. Kinetic Pro (Wgroupe, France) was
an
used for pharmacokinetic analysis. Kidney injury was assessed with neutrophil gelatinase associated lipocalin (NGAL).
M
Results: The mean total burn surface area was 36 ± 11%. The average plasma concentration of sevoflurane was 70.4 ± 37.5 mg.L-1 in burns and 57.2 ± 28.1 mg.L-1 in controls at the end
d
of the procedure (p = 0.58). The volume of distribution was higher (46.8 ± 7.2 vs 22.2 ± 2.50 l,
Ac ce pt e
p<0.001), and the drug half-life longer in burns (1.19 ± 0.28 hours vs 0.65 ± 0.04 h, p<0.0001). Free metabolite HFIP was higher in burns. Plasma fluoride was not different between burns and controls. NGAL didn’t rise after procedures. Conclusion: We observed an increased volume of distribution, slower elimination rate, and altered metabolism of sevoflurane in burn patients compared to controls. Repeated use for procedural sedation in burn patients needs further evaluation. No renal toxicity was detected. Trial registry number: ClinicalTrials.gov Identifier NCT02048683 Key words: sedation, intensive care, distribution volume, metabolism, burns The authors have no conflict of interest to declare related to this work. Of note, the Equipment (filters) was provided by Sedana Medical AB, Uppsala, Sweden. This study was partially supported by a grant from “Fondation des gueules cassées”. The funders had no influence on study design, conduct, data
Page 3 of 28
The study received approval of our Institutional Review Board in May 2013 (appropriate Institutional Review Board (Comité de Protection des Personnes Sud-Est VI, ClermontFerrand, France)) and from Agence Nationale de Sécurité du Médicament (ANSM) in June 2013 (No. EudraCT 2013-000960-26). The promoter was the University Hospital of (RBHP
2013
PERBET).
The
study
was
also
referenced
in
ip t
Clermont-Ferrand
ClinicalTrials.gov in January 2014 (identifier: NCT02048683). This manuscript adheres to the
cr
applicable EQUATOR guidelines.
us
Sevoflurane for procedural sedation in critically ill patients: a pharmacokinetic
an
comparative study between burn and non-burn patients.
Sebastien Perbet M.D. 1,2, Daniel Bourdeaux PharmD.,PhD. 3,4, Alexandre Lenoire M.D. 5,
M
Claire Biboulet M.D. 1, Bruno Pereira PhD. 6, Malha Sadoune 7, Benoit Plaud M.D., PhD. 5,7,8,
d
Jean-Marie Launay PhD. 7, 8, 9, Jean-Etienne Bazin M.D., PhD. 1, Valerie Sautou PharmD.,
Ac ce pt e
PhD. 3,4, Alexandre Mebazaa M.D., PhD. 5,7,8, Pascal Houze Pharm.D.10, Jean-Michel Constantin M.D., PhD. 1,2, Matthieu Legrand M.D., PhD. 5,7,8 for the PRONOBURN group
1
CHU Clermont-Ferrand, Intensive Care Unit, Department of Peri-Operative
Medicine, 63000 Clermont-Ferrand, France, and 2
Univ. Clermont 1, D2R2, EA-7281, Faculty of Medicine, 28 Place Henri Dunant,
F-63001 Clermont-Ferrand, France 3
CHU Clermont-Ferrand, Pharmacy department, 58 Rue Montalembert, 63003
Clermont-Ferrand, France
Page 4 of 28
4
Clermont University, Université d’Auvergne, EA4676C-BIOSENSS, 28 Place
Henri Dunant, 63000 Clermont-Ferrand, France 5
AP-HP, St-Louis Hospital, Department of Anaesthesiology and Critical Care and
6
ip t
Burn Unit, 1 Avenue Claude-Vellefaux, 75010 Paris, France CHU Clermont-Ferrand, Biostatistics Unit, DRCI, Gabriel-Montpied Hospital, 58
cr
Rue Montalembert, F-63003 Clermont-Ferrand, France 7
us
UMR INSERM 942, Institut National de la Santé et de la Recherche Médicale
(INSERM), Lariboisière hospital, 2 Rue Ambroise Paré, 75475 Paris Cedex 10 8
an
Univ Paris Diderot, Sorbonne Paris Cité, 1, Avenue Claude Vellefaux 75475
9
M
PARIS CEDEX 10, Paris, France
AP-HP, Department of biochemistry, Lariboisière hospital, 2 Rue Ambroise Paré,
AP-HP, St-Louis hospital, Department of Pharmacology, 1 Avenue Claude-
Ac ce pt e
10
d
75475 Paris Cedex 10, France
Vellefaux, 75010 Paris, France
Corresponding author: Pr Matthieu Legrand, Department of Anaesthesiology and Critical Care and Burn Unit, St-Louis Hospital, Assistance Publique- Hopitaux de Paris; France. email address: [email protected], Tel: +33 (0)1 42 49 43 48, Fax : + 33 (0)1 42 39 99 89
Financial disclosures: Equipment (filters) was provided by Sedana Medical AB, Uppsala, Sweden. This study was partially supported by a grant from “Fondation des gueules cassées”. The funders had no influence on study design, conduct, data analysis or in the preparation of this article.
Page 5 of 28
ip t cr
Abstract
Background: Sevoflurane has anti-inflammatory proprieties and short lasting effects making
us
it of interest for procedural sedation in critically ill patients. We evaluated the pharmacokinetics of sevoflurane and metabolites in severely ill burn patients and controls.
an
The secondary objective was to assess potential kidney injury.
M
Methods: prospective interventional study in a burn and a surgical intensive care unit; 24 mechanically ventilated critically ill patients (12 burns, 12 controls) were included. The
d
sevoflurane was administered with an expired fraction target of 2% during short-term
Ac ce pt e
procedural sedation. Plasma concentrations of sevoflurane, hexafluoroisopropanolol (HFIP) and free fluoride ions were recorded at different times. Kinetic Pro (Wgroupe, France) was used for pharmacokinetic analysis. Kidney injury was assessed with neutrophil gelatinase associated lipocalin (NGAL).
Results: The mean total burn surface area was 36 ± 11%. The average plasma concentration of sevoflurane was 70.4 ± 37.5 mg.L-1 in burns and 57.2 ± 28.1 mg.L-1 in controls at the end of the procedure (p = 0.58). The volume of distribution was higher (46.8 ± 7.2 vs 22.2 ± 2.50 l, p<0.001), and the drug half-life longer in burns (1.19 ± 0.28 hours vs 0.65 ± 0.04 h, p<0.0001). Free metabolite HFIP was higher in burns. Plasma fluoride was not different between burns and controls. NGAL didn’t rise after procedures.
Page 6 of 28
Conclusion: We observed an increased volume of distribution, slower elimination rate, and altered metabolism of sevoflurane in burn patients compared to controls. Repeated use for procedural sedation in burn patients needs further evaluation. No renal toxicity was detected. Trial registry number: ClinicalTrials.gov Identifier NCT02048683
Ac ce pt e
d
M
an
us
cr
ip t
Key words: sedation, intensive care, distribution volume, metabolism, burns
Page 7 of 28
Introduction Due to restriction in the use of continuous sedation in the critically ills, procedural sedation protocols for invasive procedures might be frequently indicated[1][2]. Invasive or painful procedures in the intensive care unit (ICU) may require sedation to improve patient comfort
ip t
and safety. However, the potential benefits of sedation should be balanced by potential risks,
cr
including prolonged sedation and the need for prolonged invasive mechanical ventilation. Burn patients require frequent dressing changes, which may require deep sedation[3]. The use
us
of sedative drugs with short half-life allows the patient to awake rapidly after the procedure and avoids unnecessary prolonged deep sedation[1]. Sevoflurane is a halogen gas with anti-
an
inflammatory proprieties and a short half-life and low accumulation in fat tissue, which makes
M
it a potentially appropriate sedative agent in this setting[4][5]. Its elimination produces metabolites as fluoride and hexafluoroisopropanolol (HFIP) with potential activity and
d
toxicity.
Ac ce pt e
The use of halogenated gas is facilitated via the ICU Anaconda® (or Anaesthetic Conserving Device (ACD), Sedana Medical, Sundbyberg, Sweden) or Mirus® (Pall Medical, Dreieich, Germany)[6]. The Anaconda® is a bulky medical device inserted between the patient and the ventilator that enables the administration and recycling of gas. This device also humidifies and warms up the inspired gases.
The pharmacokinetics of intravenous drugs can be modified in burn patients due to changes in the distribution volume and haemodynamic parameters[7][8]. However, there are no data available regarding the pharmacokinetic changes for inhaled sedative agents in these patients. Thus, the primary objective of this study was to evaluate the pharmacokinetics of sevoflurane for procedural sedation in severely ill burn patients and to compare the results with non-burn critically ill patients in the ICU. The secondary objective was to identify potential renal toxicity.
Page 8 of 28
Methods Patients The following study inclusion criteria were used: burn patients over 18 years of age, total burn surface area (TBSA) between 20 and 50%, and patients requiring mechanical ventilation with sedation for a predictable period of 2 to 4 hours in burn patients within 15 days of burn
ip t
injury. The controls were patients requiring mechanical ventilation with procedural sedation
cr
control for an expected period of 2 to 4 hours. The exclusion criteria were pregnancy, allergy to halogenated anaesthetic agents, inclusion in other protocols, personal or family history of
us
malignant hyperthermia, severe liver failure (factor V <50% or alanine transaminase> 500 IU. L-1), patients deprived of liberty or under guardianship, medication with isoniazid treatment,
an
diethyldithiocarbamate or disulfiram, or if the patient or next of kin refused to participate.
M
The patients (or next of kin for patients unable to consent) provided written informed consent. The burn patients were from the burn intensive care unit of St-Louis hospital, Paris, France.
d
The control ICU patients were included from the surgical ICU of Clermont-Ferrand hospital.
Ac ce pt e
The study received approval of our Institutional Review Board in May 2013 (appropriate Institutional Review Board (Comité de Protection des Personnes Sud-Est VI, ClermontFerrand, France)) and from the Agence Nationale de Sécurité du Médicament (ANSM) in June 2013 (No. EudraCT 2013-000960-26). The promoter was the University Hospital of Clermont-Ferrand
(RBHP
2013
PERBET).
The
study
was
also
referenced
in
ClinicalTrials.gov in January 2014 (identifier: NCT02048683). This manuscript adheres to the applicable EQUATOR guidelines.
The control patients were matched to burn patients for gender, age (+/- 5 years) and body mass index (BMI) (+/- 2.5 kg. m-2). Protocol All patients were managed using volume-controlled ventilation with protective ventilation. Both the ventilator frequency and tidal volume (6-8 mL.kg-1 of ideal body weight) remained
Page 9 of 28
unchanged during the sevoflurane wash-in and washout periods. The patients received continuous intravenous remifentanil infusion for painful procedures to obtain a behavioural pain score less than 3. The maintenance of sedation was achieved by inhalation of sevoflurane via the Anaconda® system with a fraction of 2% expired. We increased the flow of the
ip t
syringe pump from 6 to 10 ml.h-1 to progressively increase the fraction of sevoflurane from 0 to 2%. We collected blood samples at baseline and then after 5 min, 15 min, and 1 hour after
cr
detection (by the monitor) of sevoflurane (0.1%) in the breathing circuit. We also collected
us
blood samples immediately prior to stopping inhalation and then at 5, 10, 30, 60, and 120 minutes after cessation of sevoflurane administration. Additional samples were collected at 5
an
min, 10 min, 30 min, 2 h, 4 h, and 6 h after cessation of sevoflurane delivery (Figure 1). The blood was collected via a venous catheter directly into a 5-ml vacuum tube. We were careful
M
to limit residual air as previously described. Each sample was promptly centrifuged for 3 min at 5,000 rpm. We transferred a 1-ml sample of plasma to a 20-ml headspace tube (Antélia,
d
Décines-Charpieu, France) and then added 20 μl of an internal chromatographic standard
Ac ce pt e
solution (600 µg/ml chloroform in undecane). The tube was immediately hermetically sealed using a Teflon-sealed cap and frozen at -20°C until analysis the following week. All the samples were collected after the first sevoflurane administration. We noted the end-tidal CO2 and end-tidal sevoflurane concentrations at the time of each blood sample collection. The expiration volume/min was measured by the ventilator and was also recorded. The tubing running from the syringe to the ACD was empty at the commencement of infusion to ensure the complete absence of sevoflurane until the required time of delivery. The time of first detection of gas served as the start of kinetic calculations. The urinary and serum neutrophil gelatinase-associated lipocalin (NGAL) were collected before sedation and again at 24-h and 48-h after sedation. The serum creatinine values were monitored for 7 days following inclusion. Acute kidney injury (AKI) was defined according
Page 10 of 28
to the Kidney Disease Global Outcome, KDIGO criteria11. The baseline serum creatinine value was the value on the day of inclusion. Plasma Sevoflurane, HFIP, and fluoride quantitation Sevoflurane and HFIP were quantified using a gas chromatograph (Autosystem, Perkin Elmer Corp., San Jose, US) coupled to a
ip t
headspace system (HS40, Perkin Elmer Corp., San Jose, US) and a mass spectrometer (Clarus 500, Perkin Elmer Corp., San Jose, US) as previously described[9][6]. The mean values for
cr
accuracy and precision results obtained from the method validation process for sevoflurane
us
were 2.2% and 6.9%, respectively. The values for HFIP were 5.8% and 7.2%[10]. The limit of quantification was determined at 1 µg.mL-1 for the two molecules.
an
The total plasma fluoride levels were measured at 0 h, 24 h, and 48 h after the initiation of sevoflurane sedation. An ionometric detector (Sevenmulti, Mettler Toledo, Viroflay, France)
M
fitted with a fluoride specific electrode (Mettler Toledo, Viroflay, France) was used to quantify plasma fluoride levels. We then constructed a calibration curve using fluoride
d
standards ranging from 1-70 µmol.L-1. Each 4-ml volume of patient plasma was treated with
Ac ce pt e
400 microliters of TISAB activation solution (Mettler Toledo, Viroflay, France) prior to analysis. The precision and accuracy of the method were verified. The mean values for accuracy and precision results for fluoride obtained from the method validation process were 2% and 4.5%, respectively. Pharmacokinetic model
The aim of the present study was to establish a pharmacokinetic model of sevoflurane used with the ACD for sedation in severely ill burn patients without hepatic or renal failure. We then compared the pharmacokinetic parameters with non-burn critically ill patients. The pharmacokinetic analysis was performed with KineticPro software (Wgroupe, Pommiers la Placette, France)[6][11][12]. The observed profile (time versus observed concentrations curve) was fitted by compartmental models involving one to three compartments. As
Page 11 of 28
previously observed, the best fit was obtained by the 2-compartment model (the 3compartment was over-parameterized, and the 1-compartment model was mis-specified)[13]. The terms y1 and y2 represent the concentrations in the central and peripheral compartments, respectively. Vdis is the volume of distribution of the central compartment, and ke is the
ip t
elimination rate. The terms k1-2 and k2-1 show the exchange rates with the peripheral compartment. These parameters were estimated by nonlinear regression of y1 to the observed
cr
profile. The log-linear regression of the terminal portion of the blood concentration-time
us
curve was used to estimate the half-life of sevoflurane (T1/2). The total clearance (Cl=ke. Vdis) was also calculated.
an
We recorded sevoflurane syringe flow over time and calculated the total amount of sevoflurane administered to evaluate the amount of sevoflurane effectively absorbed. A 10%
M
leakage of sevoflurane via the ACD was effectively assumed[9]. The samples were collected
Statistical analysis
d
until 6 hours after cessation of sevoflurane delivery.
Ac ce pt e
The sample size estimation was difficult because both the between and within patient variability were unknown. However, we conducted (a priori) simulations to determine whether the sample size was adequate to allow changes to be detected over time and to estimate an intraclass correlation coefficient (ICC). We considered the results obtained for normal weight patients with standard half-life (2.84 with a standard deviation of 0.37) to select a sample size of 12 patients in each group. For 9 time points, such sample size allow to highlight a minimal difference between groups equals 0.43 with a two-sided type I error at 5%, a statistical power greater than 80% and an ICC greater than 0.8 [14].
The statistical analysis was performed using Stata software (version 13, StataCorp, College Station, US). The tests were two-sided, and the type I error was set at α=0.05. The baseline
Page 12 of 28
characteristics are presented as the mean (standard deviation (SD)) or the median [interquartile range] for continuous data (assumption of normality assessed by using the Shapiro-Wilk test). The data are shown as the number of patients and associated percentages for categorical parameters. All quantitative data were compared between groups (Burns vs. Controls) using
ip t
Student’s t-test or Mann-Whitney test when the t-test assumptions were not satisfied. The assumption of homoscedasticity was studied using the Fisher-Snedecor’s test. The
cr
comparisons between groups for categorical data were performed using either Chi-squared or
us
Fisher’s exact tests. We accounted for the between and within subject variability (due to several repeated measures for a same patient) using random-effects models (linear for
an
quantitative dependant outcome, with logarithmic transformation to achieve normality when appropriate) rather than usual statistical tests that are not appropriate because the hypothesis
M
of independent data was not verified. The patients were studied as random effects, and then
Ac ce pt e
d
the following fixed effects were analysed: group, time point evaluation, and interactions.
Page 13 of 28
Results Patients This study enrolled 24 patients (12 burns, 12 controls) between December 2013 and August 2015. In the burn patients, the TBSA was 36 ±11 %. The reasons for ICU admission in the
ip t
control subjects were related to surgery in five patients and were medical in seven patients.
cr
The control and burn patients were not different regarding gender, age, and BMI (Table 1). The controls had non-significantly higher Simplified Acute Physiology Score II (SAPSII) (45
us
± 12 versus 35 ± 20), p = 0.27), higher serum creatinine, and impaired renal function (92 ± 30 μmol.L-1 vs. 71 ± 39 μmol.L-1, p = 0.18) compared to the burn cases. One patient in each
an
group died while in the ICU.
M
Pharmacokinetics of sevoflurane
The plateau plasma concentrations of sevoflurane were not significantly different between
d
groups. However, the drug concentrations were higher in burn cases during the washout phase.
Ac ce pt e
The concentration increased rapidly to 73 ± 38 mg.L-1 and 57 ± 28 mg.L-1 in burn cases and controls at the end of the administration of sevoflurane (p=0.27) (Figure 2), respectively. The rapid plasma sevoflurane washout was associated with a rapid decrease of sevoflurane concentration after the end of sevoflurane inhalation (Figure 2). During the procedure, the subjects inhaled 61.01 ± 26.62 mg of sevoflurane in 181 ± 62 minutes and 59.33 ± 19.86 mg sevoflurane in 182 ± 65 minutes in burn cases and controls, respectively (p = 0.51). The expired fractions of sevoflurane were similar between groups. Table 2 describes the 2-compartment pharmacokinetic model constants calculated by the pharmacokinetic software. The burn cases had higher Vdis, lower ke, longer k1-2 and k2-1, longer T1/2 and lower Cl.
Page 14 of 28
The concentration of sevoflurane was correlated with the expired fraction of sevoflurane both in burn cases and controls (p < 0.01). The evolution of ventilation parameters is described in supplementary file 1. Plasma hexafluoroisopropanolol and fluoride
ip t
The level of free HFIP was significantly higher in burn cases than controls (p=0.02, Figure 3). The maximum values of plasma concentration of free HFIP were 5.05 ± 2.26 mg.mL-1 at 1
cr
hour and 1.63 ± 1.19 mg.mL-1 at 30 minutes after discontinuation of sedation sevoflurane in
us
burn cases and controls, respectively. The mean plasma fluoride values peaked 24 h after the end of inhalation, with a maximum of 22 ± 14 mg.mL-1 in burn cases and 38 ± 34 mg.mL-1 in
an
controls (p=0.16). Renal function and injury
M
The controls had higher serum creatinine levels than the burn cases throughout the 7-day observational period following inclusion. However, there were no significant changes
d
observed within groups (Supplementary file 2). The renal function was not significantly
Ac ce pt e
altered in the days following sedation with sevoflurane in burn cases. One burn patient (8%) and four (33%) controls met the criteria for AKI during the 7 days following inclusion. The cause of AKI in the burn patient was attributed to septic shock. Interestingly, we did not observe significant elevation of plasma and urine NGAL after inclusion in both groups. The urine and plasma NGAL remained < 150 ng.mL-1 in all patients except one (maximum 154 ng.mL-1) (Figure 4).
Page 15 of 28
Discussion In this study, we compared the pharmacokinetics of inhaled sevoflurane for short-term procedural sedation in severely ill burn patients and non-burn critically ill patients. We observed a rapid rise in sevoflurane plasma concentration after initiation of sedation in both
ip t
groups. However, burn patients showed higher volume of distribution and lower clearance.
cr
Furthermore, the higher HFIP levels suggest there is altered hepatic metabolism in burn
the stable plasma and urine NGAL level after procedures.
us
patients. Finally, no evidence of kidney injury could be observed in this cohort as shown by
Prolonged deep sedation is no longer recommended in critically ill patients[15]. However,
an
short-term deep sedation may be used to improve patient comfort and safety in painful or
M
invasive procedures. Burn patients require iterative procedural sedation for dressing changes. The inhaled sevoflurane appears to have appropriate pharmacokinetics and allows for short-
d
term sedation with theoretical rapid clearance[6]. The pharmacokinetics of several
Ac ce pt e
intravenous medications is altered in burn patients with increased distribution volume but no pharmacokinetics data are available regarding inhaled sedation with sevoflurane. Despite the pharmacokinetic properties that make inhaled sevoflurane suitable for short-term sedation (rapid increase of plasma concentration and rapid plasma clearance), there were significant differences compared to non-burn critically ill patients matched for age, gender, and BMI. The two important features were increased plasma sevoflurane concentration at the end of the procedure, despite similar inhaled doses and duration of sedation. The reasons for the higher concentration might be related to higher cardiac output and lung perfusion increasing the contact between alveolar gases and blood in burn patients compared to controls and primary injury to the lungs and airways from direct thermal inhalation, or widespread systemic inflammation[16]. Second, clearance of sevoflurane was lower with higher half-life in burn cases, which could be associated with a renal or non-renal clearance modification.
Page 16 of 28
However, there was a rapid decrease in plasma concentration observed. This decrease was associated with a significant increase in distribution volume generally observed in burn patients with increased capillary permeability and fluid extravasation[17]. These explanations are speculative because we could not determine the blood/gas coefficients for sevoflurane in
ip t
the burn and control patients. The concentration of free HFIP was higher in the burn patients. The reason for this
cr
observation could be modified by hypoalbuminaemia related to the loss of protein-rich fluid
us
via the burn wound, a decrease in constitutive hepatic protein synthesis, and an increase in catabolism (with an increase of the unbound (free) fraction of a drug) and by hepatic
an
metabolism of sevoflurane with lower glucuroconjugation and/or overall decrease in renal elimination[18]. The possibility of altered renal clearance is unlikely because the renal
M
function was normal in the burn patients in this cohort. Conversely, hepatic metabolism has previously been found to be impaired after burn injury[19][20][18]. Further studies should
d
explore sevoflurane hepatic metabolism in burn cases for longer periods and/or following
Ac ce pt e
iterative sedation with sevoflurane and exclude the potential accumulation of its metabolites. A decreased clearance may lead to accumulation of sevoflurane during prolonged and/or iterative administration in burn patients and raises concerns regarding safety in this specific population.
In our study, there was no significant impact of sevoflurane on kidney injury. Although several patients did develop AKI, ICU patients have many factors associated with a risk of AKI (e.g., sepsis, trauma, hemodynamic, nephrotoxic, etc.). In this study, we measured plasma and urine NGAL. NGAL is a highly sensitive biomarker of AKI after sevoflurane administration[21]. NGAL did not significantly rise within 48 hours after sevoflurane administration. We used a cut-off of 150 ng.mL-1, which has been reported as highly sensitive to AKI,[22][23] and found all patients except one (with a peak at 154 ng.mL-1) did
Page 17 of 28
not reach this cut-off. Therefore, it is unlikely that sevoflurane promoted the development of AKI in these cohorts. Finally, the fluoride values remained low. In contrast to the renal metabolism of the methoxyflurane that may have resulted in renal intraparenchymal concentrations above 50
ip t
µmol.L-1 [24], the hepatic metabolism of the sevoflurane may explain the renal toxicity of
cr
fluoride when high fluoride levels are not observed. Renal toxicity was however not observed in our study.
us
Sevoflurane therefore appears to have slight different metabolism in burn and non-burn patients but the pharmacokinetics however appear to be safe – especially with no detection of
an
kidney injury – using transient procedural sedation in both populations. Based on this results,
M
we feel that sevoflurane could be used safely in both population in this setting for procedural sedation but that safety of continuous use or multiple procedural sedation with <48h between
Ac ce pt e
d
administration would require additional investigations.
Our study has several limitations. Our data shows a wide range of values for the plasma sevoflurane concentration and its metabolite, which suggests there is inter-individual variability of sevoflurane pharmacokinetics. The intra-individual variability was not explored in this study. Although this was a multicentre study, all burn cases and controls were included in a single centre. Finally, only short-term sedation with samples until only 6 hours after cessation of sevoflurane delivery were studied, and the results could not be extrapolated to long-term sedation or iterative procedures.
To conclude, we confirm rapid clearance of sevoflurane in non-burn critically ill patients during procedural sedation. We however observed altered metabolism and accumulation of HFIP, larger volume of distribution, and lower clearances of sevoflurane in burn patients.
Page 18 of 28
These differences must be considered while using sevoflurane for sedation in burn patients. Impact of iterative procedural sedations on sevoflurane and its metabolites clearance should
Ac ce pt e
d
M
an
us
cr
ip t
be explored.
Page 19 of 28
Acknowledgements: We thank the nurses, residents, and physicians who provided care for the study patients and collected data. We also thank Dr Hélène Sauvageon (pharmacy, St-Louis Hospital, Paris,
Ac ce pt e
d
M
an
us
cr
ip t
France) for fluoride dosages.
Page 20 of 28
References
Ac ce pt e
d
M
an
us
cr
ip t
[1] Barr J, Fraser GL, Puntillo K, Ely EW, Gélinas C, Dasta JF, et al. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med 2013;41:263–306. doi:10.1097/CCM.0b013e3182783b72. [2] Hughes CG, Girard TD, Pandharipande PP. Daily sedation interruption versus targeted light sedation strategies in ICU patients. Crit Care Med 2013;41:S39–45. doi:10.1097/CCM.0b013e3182a168c5. [3] Gregoretti C, Decaroli D, Piacevoli Q, Mistretta A, Barzaghi N, Luxardo N, et al. Analgo-sedation of patients with burns outside the operating room. Drugs 2008;68:2427–43. [4] Jabaudon M, Boucher P, Imhoff E, Chabanne R, Faure J-S, Roszyk L, et al. Sevoflurane for Sedation in Acute Respiratory Distress Syndrome. A Randomized Controlled Pilot Study. Am J Respir Crit Care Med 2017;195:792–800. doi:10.1164/rccm.2016040686OC. [5] Chabanne R, Perbet S, Futier E, Ben Said NA, Jaber S, Bazin J-E, et al. Impact of the anesthetic conserving device on respiratory parameters and work of breathing in critically ill patients under light sedation with sevoflurane. Anesthesiology 2014;121:808–16. doi:10.1097/ALN.0000000000000394. [6] Perbet S, Bourdeaux D, Sautou V, Pereira B, Chabanne R, Constantin JM, et al. A pharmacokinetic study of 48-hour sevoflurane inhalation using a disposable delivery system (AnaConDa®) in ICU patients. Minerva Anestesiol 2014;80:655–65. [7] Jaehde U, Sörgel F. Clinical pharmacokinetics in patients with burns. Clin Pharmacokinet 1995;29:15–28. doi:10.2165/00003088-199529010-00003. [8] Mesnil M, Capdevila X, Bringuier S, Trine P-O, Falquet Y, Charbit J, et al. Long-term sedation in intensive care unit: a randomized comparison between inhaled sevoflurane and intravenous propofol or midazolam. Intensive Care Med 2011;37:933–41. doi:10.1007/s00134-011-2187-3. [9] Meiser A, Bellgardt M, Belda J, Röhm K, Laubenthal H, Sirtl C. Technical performance and reflection capacity of the anaesthetic conserving device--a bench study with isoflurane and sevoflurane. J Clin Monit Comput 2009;23:11–9. doi:10.1007/s10877-0089158-4. [10] Bourdeaux D, Sautou-Miranda V, Montagner A, Perbet S, Constantin JM, Bazin J-E, et al. Simple assay of plasma sevoflurane and its metabolite hexafluoroisopropanol by headspace GC-MS. J Chromatogr B Analyt Technol Biomed Life Sci 2010;878:45–50. doi:10.1016/j.jchromb.2009.11.018. [11] Iliadis A, Brown AC, Huggins ML. APIS: a software for model identification, simulation and dosage regimen calculations in clinical and experimental pharmacokinetics. Comput Methods Programs Biomed 1992;38:227–39. [12] Petricoul O, Claret L, Barbolosi D, Iliadis A, Puozzo C. Information tools for exploratory data analysis in population pharmacokinetics. J Pharmacokinet Pharmacodyn 2001;28:577–99. [13] Enlund M, Kietzmann D, Bouillon T, Züchner K, Meineke I. Population pharmacokinetics of sevoflurane in conjunction with the AnaConDa: toward target-controlled infusion of volatiles into the breathing system. Acta Anaesthesiol Scand 2008;52:553–60. doi:10.1111/j.1399-6576.2008.01579.x. [14] Ogungbenro K, Aarons L. How many subjects are necessary for population pharmacokinetic experiments? Confidence interval approach. Eur J Clin Pharmacol 2008;64:705–13. doi:10.1007/s00228-008-0493-7.
Page 21 of 28
Ac ce pt e
d
M
an
us
cr
ip t
[15] Barr J, Pandharipande PP. The pain, agitation, and delirium care bundle: synergistic benefits of implementing the 2013 Pain, Agitation, and Delirium Guidelines in an integrated and interdisciplinary fashion. Crit Care Med 2013;41:S99–115. doi:10.1097/CCM.0b013e3182a16ff0. [16] Enkhbaatar P, Traber DL. Pathophysiology of acute lung injury in combined burn and smoke inhalation injury. Clin Sci Lond Engl 1979 2004;107:137–43. doi:10.1042/CS20040135. [17] Udy AA, Roberts JA, Lipman J, Blot S. The effects of major burn related pathophysiological changes on the pharmacokinetics and pharmacodynamics of drug use: An appraisal utilizing antibiotics. Adv Drug Deliv Rev 2018;123:65–74. doi:10.1016/j.addr.2017.09.019. [18] Yogaratnam D, Ditch K, Medeiros K, Miller MA, Smith BS. The Impact of Liver and Renal Dysfunction on the Pharmacokinetics and Pharmacodynamics of Sedative and Analgesic Drugs in Critically Ill Adult Patients. Crit Care Nurs Clin North Am 2016;28:183– 94. doi:10.1016/j.cnc.2016.02.009. [19] Pereira CT, Herndon DN. The pharmacologic modulation of the hypermetabolic response to burns. Adv Surg 2005;39:245–61. [20] Finnerty CC, Jeschke MG, Qian W-J, Kaushal A, Xiao W, Liu T, et al. Determination of burn patient outcome by large-scale quantitative discovery proteomics. Crit Care Med 2013;41:1421–34. doi:10.1097/CCM.0b013e31827c072e. [21] Ronco C, Legrand M, Goldstein SL, Hur M, Tran N, Howell EC, et al. Neutrophil gelatinase-associated lipocalin: ready for routine clinical use? An international perspective. Blood Purif 2014;37:271–85. doi:10.1159/000360689. [22] Zarbock A, Kellum JA, Schmidt C, Van Aken H, Wempe C, Pavenstädt H, et al. Effect of Early vs Delayed Initiation of Renal Replacement Therapy on Mortality in Critically Ill Patients With Acute Kidney Injury: The ELAIN Randomized Clinical Trial. JAMA 2016;315:2190–9. doi:10.1001/jama.2016.5828. [23] Legrand M, Mebazaa A, Ronco C, Januzzi JL. When cardiac failure, kidney dysfunction, and kidney injury intersect in acute conditions: the case of cardiorenal syndrome. Crit Care Med 2014;42:2109–17. doi:10.1097/CCM.0000000000000404. [24] Kharasch ED, Hankins DC, Thummel KE. Human kidney methoxyflurane and sevoflurane metabolism. Intrarenal fluoride production as a possible mechanism of methoxyflurane nephrotoxicity. Anesthesiology 1995;82:689–99.
Page 22 of 28
Tables Table 1. Patient characteristics. BMI, body mass index; TBSA, total body surface area; SAPS, simplified acute physiology score; GFR, glomerular filtration rate. Data expressed in mean (SD), (SD, standard deviation). Controls (n=12) 10 (83%) 55 (17) 71,75 (11,87) 1,70 (0,08) 24,66 (3,31) NA 42,83 (13,78) 10,95 (3,52) 92,18 (30,28) 81,71 (29,1) 182,25 (64,92)
an
us
cr
ip t
Burns (n= 12) 8 (67%) 49 (17) 80,25 (12,44) 1,74 (0,12) 26,27 (3,07) 36,13 (11,11) 34,50 (19,84) 7,84 (4,80) 71,33 (39,88) 133,83 (76,48) 181,33 (62,08)
Male sex - n (%) Age (years) Weight (kg)Height (m) - mean (SD) BMI (kg.m-2) TBSA (%) SAPS 2 Plasma urea (mmol.L-1) at inclusion Plasma creatinine (µmol.L-1) at inclusion GFR (ml.min-1) Sedation duration (min)
Ke (h-1)
46.80 (7.20) 22.20 (2.50) <0.0001
5.95 (0.92) 14.70 (1.61) <0.0001
K1-2 (h-1)
K2-1 (h-1)
T½ (h)
Clpl (l/h)
RE
2.44 (0.73) 4.80 (0.94) <0.0001
0.85 (0.24) 1.44 (0.12) <0.0001
1.19 (0.28) 0.65 (0.04) <0.0001
279.00 (61.10) 327.00 (51.90) 0.04
13.98% 7.55%
Ac ce pt e
Burns (n=12) Controls (n=12) P
Vdis (l)
d
M
Table 2. Pharmacokinetics of sevoflurane in burn cases and controls. The terms; y1 and y2 are the concentrations in the central and peripheral compartments, respectively; Vdis, the volume of distribution of the central compartment (in liters); ke, the elimination rate; k1-2 and k2-1, the exchange rates with the peripheral compartment; T1/2, the half-life of sevoflurane; Clpl, total plasma clearance; RE, residual error. Data expressed in mean (SD), (SD, standard deviation).
Page 23 of 28
Figures legends Figure 1: Study protocol (T0 = start of the sedation, T4 = end of sedation). Figure 2: Expired fraction (Exp, %) and plasma concentrations (mg.L-1) of sevoflurane in control and burn patients.
ip t
Figure 3: Plasma hexafluoroisopropanolol (HFIP) concentration (mg.L-1) in control and burn patients.
cr
Figure 4: Urine and plasma NGAL values (ng. L-1) in control and burn patients (values
Ac ce pt e
d
M
an
us
expressed in median and interquartilerange).
Page 24 of 28
Glossary
AKI, Acute kidney injury ACD, Anaesthetic Conserving Device
ip t
BMI, body mass index
cr
HFIP, hexafluoroisopropanolol NGAL, neutrophil gelatinase associated lipocalin
us
SAPSII, Simplified Acute Physiology Score II
an
TBSA, Total body surface area
d
M
Table 1. Patient characteristics. BMI, body mass index; TBSA, total body surface area; SAPS, simplified acute physiology score; GFR, glomerular filtration rate. Data expressed in mean (SD), (SD, standard deviation).
Ac ce pt e
Male sex - n (%) Age (years) Weight (kg)Height (m) - mean (SD) BMI (kg.m-2) TBSA (%) SAPS 2 Plasma urea (mmol.L-1) at inclusion Plasma creatinine (µmol.L-1) at inclusion GFR (ml.min-1) Sedation duration (min)
Burns (n= 12) 8 (67%) 49 (17) 80,25 (12,44) 1,74 (0,12) 26,27 (3,07) 36,13 (11,11) 34,50 (19,84) 7,84 (4,80) 71,33 (39,88) 133,83 (76,48) 181,33 (62,08)
Controls (n=12) 10 (83%) 55 (17) 71,75 (11,87) 1,70 (0,08) 24,66 (3,31) NA 42,83 (13,78) 10,95 (3,52) 92,18 (30,28) 81,71 (29,1) 182,25 (64,92)
Table 2. Pharmacokinetics of sevoflurane in burn cases and controls. The terms; y1 and y2 are the concentrations in the central and peripheral compartments, respectively; Vdis, the volume of distribution of the central compartment (in liters); ke, the elimination rate; k1-2 and k2-1, the exchange rates with the peripheral compartment; T1/2, the half-life of sevoflurane; Clpl, total plasma clearance; RE, residual error. Data expressed in mean (SD), (SD, standard deviation).
Vdis (l)
Ke (h-1)
K1-2 (h-1)
K2-1 (h-1)
T½ (h)
Clpl (l/h)
RE
Page 25 of 28
2.44 (0.73) 0.85 (0.24) 1.19 (0.28) 4.80 (0.94) 1.44 (0.12) 0.65 (0.04) <0.0001 <0.0001 <0.0001
279.00 (61.10) 327.00 (51.90) 0.04
13.98% 7.55%
d
M
an
us
cr
ip t
46.80 (7.20) 5.95 (0.92) 22.20 (2.50) 14.70 (1.61) <0.0001 <0.0001
Ac ce pt e
Burns (n=12) Controls (n=12) P
Page 26 of 28
Page 27 of 28
d
Ac ce pt e us
an
M
cr
ip t
2
1
0
D
D
0
0
2
urine NGAL (ng.mL-1)
ed controls burns
D
D
60
pt
ce
Ac
20
1
0
40
D
D
plasma NGAL (ng.mL-1) 80 100
80
controls burns
60
40
20
Page 28 of 28