| ||||||||||||||
|
|
|||||||||||||





*Department of Anesthesiology, Duke University Medical Center, Durham, North Carolina;
Departments of Anesthesiology and Pediatrics, University of Pittsburgh School of Medicine, Childrens Hospital of Pittsburgh, Pittsburgh, Pennsylvania;
Department of Anesthesiology, Harvard Medical School, and Boston Childrens Hospital, Boston, Massachusetts; and
Glaxo Wellcome, Inc., Research Triangle Park, North Carolina
Address correspondence and reprint requests to Peter J. Davis, MD, Professor of Anesthesiology and Pediatrics, Department of Anesthesiology, Childrens Hospital of Pittsburgh, 3705 Fifth Ave., Pittsburgh, PA 15213-2583. Address e-mail to davis{at}smtp.anes.upmc.edu
| Abstract |
|---|
|
|
|---|
2 mo), older infants (>2 mo to <2 yr), young children (2 to <7 yr), older children (7 to <13 yr), adolescents (13 to <16 yr), and young adults (16 to <18 yr). Arterial blood samples were collected and analyzed by mass spectroscopy to determine remifentanil pharmacokinetic profiles. Hemodynamic measurements for remifentanils effect were made after the infusion. Methods of statistical analysis included analysis of variance and linear regression, with significance at P
0.05. Complete remifentanil pharmacokinetic data were obtained from 34 patients. The volume of distribution was largest in the infants <2 mo (mean, 452 mL/kg) and decreased to means of 223 to 308 mL/kg in the older patients. There was a more rapid clearance in the infants <2 mo of age (90 mL · kg-1 · min-1) and infants 2 mo to 2 yr (92 mL · kg-1 · min-1) than in the other groups (means, 46 to 76 mL · kg-1 · min-1). The half-life was similar in all age groups, with means of 3.4 to 5.7 min. Seven subjects (17%) developed hypotension related to the remifentanil bolus. Remifentanil showed an extremely rapid elimination similar to that in adults. The fast clearance rates observed in neonates and infants, as well as the lack of age-related changes in half-life, are in sharp contrast to the pharmacokinetic profile of other opioids. Remifentanil in a bolus dose of 5 µg/kg may cause hypotension in anesthetized children. IMPLICATIONS: The pharmacokinetics of remifentanil were studied in children from birth to 18 yr. Remifentanil was found to have age-related changes in clearance and volume of distribution, but not half-life. The increased clearance observed in young infants is in contrast to other opioids.
| Introduction |
|---|
|
|
|---|
The pharmacokinetics of this rapidly metabolized drug in children have not yet been reported, and because of its unique esterase metabolism, extrapolation from reported adult pharmacokinetic data may be inappropriate. The aim of our study was to determine the pharmacokinetics of remifentanil and its metabolite (GR90291) and to describe the hemodynamic response after a single bolus injection in pediatric patients from ages 0 to 18 yr.
| Methods |
|---|
|
|
|---|
2 mo), older infants (>2 mo to <2 yr), young children (2 to <7 yr), older children (7 to <13 yr), adolescents (13 to <16 yr), and young adults (16 to <18 yr). Children were excluded from consideration if there was clinical or laboratory evidence of hepatic or renal disease, known hypersensitivity to opioids, or a history of medications such as esmolol, fentanyl, or alfentanil that could interfere with the metabolism or disposition of remifentanil. Children older than 2 yr whose procedure included the use of cardiopulmonary bypass (CPB) were excluded from the study. However, in children younger than 2 yr, blood sampling time was limited to 1 h, and procedures involving CPB were included provided that CPB was initiated after the 1-h sample. All the patients were premedicated with oral, nasal, or IV midazolam or oral diazepam. Anesthesia was induced IV with sodium thiopental or by mask with nitrous oxide and oxygen, with incremental increasing concentrations of halothane. After insertion of a venous catheter, patients received a neuromuscular blocking drug to facilitate tracheal intubation, control ventilation, and prevent chest wall rigidity. Neuromuscular blocking drugs included pancuronium (25 of 42 patients), rocuronium (9 of 42), vecuronium (5 of 42), atracurium (1 of 42), cisatracurium (1 of 42), and succinylcholine (1 of 42). Anesthesia was maintained with 60% nitrous oxide in oxygen and isoflurane. Atropine was given at the discretion of the anesthesiologist. Ventilation was controlled to maintain an end-tidal CO2 level of 3040 mm Hg. Body temperature was maintained at >35°C. Monitoring consisted of electrocardiography (standard lead II), noninvasive and invasive measurements of arterial pressure, end-tidal CO2 analysis, and pulse oximetry.
After an arterial catheter was inserted and hemodynamic variables were stable (three consecutive noninvasive blood pressure measurements and heart rates within 10% of baseline as measured over a period of 5 min), remifentanil 5 µg/kg was administered into a peripheral vein over 1 min via Baxter (McGaw Park, IL) infusion pump Model AS40A. Each patients remifentanil dose was prepared in a total volume of 5 mL. Blood pressure and heart rate were measured continuously and recorded at predose, at the end of the bolus infusion (1 min), and at 2, 3, 4, 5, 10, 20, 30, 45, and 60 min after the start of the bolus infusion.
To minimize the volume of blood withdrawn in both infant groups, arterial blood samples of 0.5 mL were collected at baseline and at 1, 2, 3, 5, 10, 20, 30, 45, and 60 min after the start of the infusion. Remifentanil metabolite GR90291 was not determined in this young age group. In subjects 2 to 18 yr of age, 3-mL arterial blood samples were collected at baseline and at 1, 2, 3, 5, 10, 20, 30, 45, 60, 120, 180, and 240 min after the start of the infusion. Degradation of the remifentanil was inhibited by immediately mixing the blood sample with 50% citric acid. There was then liquid-liquid extraction of the denatured blood samples with methylene chloride and separation of the extracted blood and organic phases. The lower organic layer and the blood layer were frozen for later analysis of remifentanil and GR90291 concentrations, respectively, by validated GC-High Resolution Mass Spectrometry-Selected Ion Monitoring (Triangle Labs, Research Triangle Park, NC) (7,8). The GC-High Resolution Mass Spectroscopy assay is capable of detecting remifentanil to 0.1 ng/mL and the metabolite to 0.5 ng/mL, with the upper limit of quantitation at 100 ng/mL for each. The interassay coefficients of variation for remifentanil and the metabolite were 9.3% and 13%, respectively.
Model-independent pharmacokinetic analysis was determined by using WinNonlin Professional V1.5 (SCI, Cary, NC). The peak plasma concentration, volume of distribution at steady state (VDss), CL, and t1/2 were measured or calculated by using standard methods. The observed peak plasma concentration was obtained directly from the individual plasma concentration/time data. The area under the curve of concentration versus time was calculated for each subject by using the linear trapezoidal rule before the achievement of the maximum concentration and the log-linear trapezoidal rule thereafter and by using standard algorithms from the software. The elimination rate constant was estimated from three or more points on the terminal elimination phase. These points were selected by the software and accepted, if appropriate, or manually changed where the program selected inappropriate data for the determination of elimination rate constant. CL and VDss were calculated with standard moment methods as implemented in the software.
Statistical analysis was performed with SAS software version 6.12 (SAS Institute, Inc.). Analysis of variance was used for comparison between age groups. A P value of
0.05 was considered statistically significant. On the basis of estimates of variability in CL and VDss from adult studies, eight patients per group would be sufficient to detect a 30% difference in CL and a 40% change in VDss (80% power,
= 0.05) between groups. In addition, SAS PROC REG was used for the linear regression analysis by using age as the independent variable and the pharmacokinetic variable as the dependent variable. Linear regression analyses were performed to correlate changes in age with CL, VDss, and t1/2, and these were plotted with Microsoft Excel 97, version SR-1 (Microsoft Corp., Redmond, WA).
| Results |
|---|
|
|
|---|
The demographics, surgical procedures, and pharmacokinetic variables for the individual patients are presented in Table 1. In eight of the patients, blood samples were insufficient to construct plasma decay curves. Of these eight patients, two had sampling difficulties with the arterial catheter; two patients had an insufficient number of measurable concentrations of remifentanil; two patients, one aged 8 wk and one aged 2 yr, had uninterpretable variable estimates; and two had indeterminate concentrations secondary to technical difficulties with the assay. In the remaining 34 patients, adequate blood samples were available to determine remifentanils pharmacokinetic profile. The plasma concentrations for each patient in each age group are presented in Figures 16 for remifentanil. The concentration of remifentanil in arterial blood peaked at the end of the 1-min infusion and declined rapidly. The pharmacokinetic variables and coefficients of variation of remifentanil for the age groups are displayed in Table 2. The ratio of the area under the curve (AUC) last data point to AUC infinity was 92% ± 6%. Infants, particularly neonates, had an increased CL and increased VDss. The t1/2 was similar among the groups. Twenty of the 34 patients had sampling for measurements of GR90291. The pharmacokinetic variables are presented in Table 3. The concentration of the metabolite peaked at 10 to 30 min after the infusion and declined more gradually than that of remifentanil. GR90291 kinetics were similar in all the older age groups.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Perhaps even more noteworthy is the predictability of remifentanils pharmacokinetics. All other opioids have highly variable pharmacokinetic variables in the neonatal period. This large pharmacokinetic variability of drugs dependent on hepatic biotransformation and elimination in the neonatal period is not surprising in view of the rapid changes that occur with both anatomic growth and physiologic maturation during the first few months of life. We chose the coefficient of variation (CV) as a means of quantifying variability (the CV is the ratio of SD to mean expressed as a percentage). In the neonatal period, the values reported for the CV of drug CL for fentanyl, alfentanil, and sufentanil are 91%, 109%, and 91%, respectively (1315). These values are nearly threefold larger than the value of 36.7% that we determined for remifentanil. Therefore, if variability in remifentanils effect mimics its pharmacokinetic variability, remifentanil should be a very predictable drug, even in the neonatal population.
The primary metabolite of remifentanil is the remifentanil acid GR90291, which is primarily eliminated in the urine. Because of ethical concerns regarding excessive blood sampling, we did not study the kinetics of the metabolite in patients younger than 2 years of age. Thus, our data analysis was restricted to 20 older children. In these patients, the t1/2 of GR90291 ranged from 55 to 105 minutes, and the mean AUCinf ratio (GR90291/remifentanil) ranged from 6.0 to 10.5 (AUCinf is the area under the blood concentration/time curve from time 0 to infinity in ng · min-1 · mL-1). The mean AUCinf ratio (GR90291/remifentanil) suggests that at steady state, the metabolite should be approximately 6 to 10 times the steady-state concentration of remifentanil. In view of the relatively low potency (1:4600) of the metabolite compared with remifentanil, GR90291 contributes very little to the opioid effects of remifentanil in animals (17). Because of our small sample size, comparisons of the metabolite kinetic profile in children to published results in adult surgical patients (t1/2, 88 to 137 minutes; AUCinf ratio of GR90291 to AUCinf of remifentanil, 12) is difficult. In addition, we did not determine the pharmacokinetic profile of the metabolite in infants and neonates and therefore eliminate age-related changes in metabolite elimination. Although age-related changes occur with renal function and these changes may effect GR90291 elimination, nonetheless the metabolites relative low potency should yield a negligible clinical effect.
The hemodynamic responses to remifentanil (decrease in heart rate and decrease in blood pressure) were similar to those seen with other µ-acting opioids. The frequent incidence of drug-related hypotension (17%) reflects the large dose that was administered (5 µg/kg) over 1 min. This dose was selected so that the pharmacokinetic profile of the drug could be better delineated and far exceeds the recommended clinical dose of 12.5 µg · kg-1 · min-1 (18,19). Despite this larger dose, there was little change in heart rate, and this may have been caused by the variability in the use of atropine and vagolytic neuromuscular blockers.
Four issues of our study could be raised. First, we included patients with congenital heart disease in the youngest age groups and not in the older groups. Thus, it is conceivable that intracardiac shunts, rather than age, influenced the patients pharmacokinetic profile. Because arterial catheters are infrequently needed for noncardiac surgery in neonates, we chose to include infants undergoing cardiac surgical repair to aid patient enrollment. Similar to what was found in the infants in our study, Koren et al. (20), in a study of children with tetralogy of Fallot, noted that fentanyl CL was highest in infants. In children undergoing atrial septal defect repairs in which remifentanil pharmacokinetics were determined both before and after CPB, Davis et al. (21) noted that remifentanil kinetics are minimally affected when compared with other opioids.
Another criticism of our study is that we did not correlate remifentanil pharmacokinetics to weight or body surface area, but rather to age. Although in adults Egan et al. (22) suggested that remifentanil dosage should be based on ideal body weight because dosing based on total body weight resulted in excessively large remifentanil concentrations in obese patients, we studied patients in whom weight was close to ideal.
The third issue that could be raised about our study design is the difference in plasma sampling times used for the various age groups. For ethical reasons, in the youngest patients we limited the total volume of blood sampled to 7 mL. Consequently, we terminated plasma sampling at 60 minutes after drug injection and, therefore, did not measure the remifentanil metabolite. Although differences in plasma sampling times can profoundly affect calculated pharmacokinetic values, in all our patients we noted no detectable plasma remifentanil concentration at 45 minutes. Thus, the differences in plasma sampling times did not affect the pharmacokinetic profiles.
The fourth issue is that the hemodynamic data after remifentanil administration may have been influenced by the choice of muscle relaxant and prior use of atropine. Although atropine and muscle relaxants with vagolytic side effects can attenuate the bradycardia associated with opioids, when we separately analyzed those patients who received atropine, pancuronium, or both, we noted no statistical difference in hemodynamic changes from those patients who did not receive atropine or pancuronium.
In summary, we observed age-related changes in the pharmacokinetic profile of remifentanil. Unlike other opioids, for which neonates have the slowest CLs, largest VDss, and longest t1/2, remifentanil has a more rapid CL and greater volume of distribution in infants <2 months old compared with older children. Remifentanils t1/2, however, does not change with age. From these data we would predict that a faster infusion rate should be used for neonates and infants than for older children, with the assumption that the target levels should be the same. In addition, the low coefficient of variability for remifentanil CL suggests that this drug should have little interpatient variability. Further studies will be needed to ascertain whether the decrease in pharmacokinetic variability is associated with a similar decrease in pharmacodynamic variability.
| Acknowledgments |
|---|
The authors acknowledge the following individuals: Annette Eddins and Sue Danfelt for their secretarial support, Annette Wilson, MS, and Sephali Chakravorti, PhD, for their technical assistance, statisticians Meredith Decker, MS, and Cameron Huffman, MS, and for editing, Lisa Cohn.
| Footnotes |
|---|
Presented in part at the meetings of the International Anesthesia Research Society, 1994; the American College of Chest Physicians, 1997; and the American Society of Anesthesiologists, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Jeleazcov, H. Ihmsen, J. Schmidt, C. Ammon, H. Schwilden, J. Schuttler, and J. Fechner Pharmacodynamic modelling of the bispectral index response to propofol-based anaesthesia during general surgery in children Br. J. Anaesth., April 1, 2008; 100(4): 509 - 516. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. E. Rigby-Jones, M. J. Priston, J. R. Sneyd, A. P. McCabe, G. I. Davis, M. A. Tooley, G. C. Thorne, and A. R. Wolf Remifentanil-midazolam sedation for paediatric patients receiving mechanical ventilation after cardiac surgery Br. J. Anaesth., August 1, 2007; 99(2): 252 - 261. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Reyntjens, L. Foubert, D. D. Wolf, G. Vanlerberghe, and E. Mortier Glycopyrrolate during sevoflurane-remifentanil-based anaesthesia for cardiac catheterization of children with congenital heart disease Br. J. Anaesth., November 1, 2005; 95(5): 680 - 684. [Abstract] [Full Text] [PDF] |
||||
![]() |
P.-A. Lonnqvist and N. S. Morton Postoperative analgesia in infants and children Br. J. Anaesth., July 1, 2005; 95(1): 59 - 68. [Full Text] [PDF] |
||||
![]() |
H. R. Glaisyer and M. R. J. Sury Recovery After Anesthesia for Short Pediatric Oncology Procedures: Propofol and Remifentanil Compared with Propofol, Nitrous Oxide, and Sevoflurane Anesth. Analg., April 1, 2005; 100(4): 959 - 963. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Chanavaz, O. Tirel, E. Wodey, J. Y. Bansard, L. Senhadji, J. C. Robert, and C. Ecoffey Haemodynamic effects of remifentanil in children with and without intravenous atropine. An echocardiographic study Br. J. Anaesth., January 1, 2005; 94(1): 74 - 79. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. K. Weale, C. A. Rogers, R. Cooper, J. Nolan, and A. R. Wolf Effect of remifentanil infusion rate on stress response to the pre-bypass phase of paediatric cardiac surgery Br. J. Anaesth., February 1, 2004; 92(2): 187 - 194. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Yumul, A. Emdadi, and N. Moradi Anesthesia for Noncardiac Surgery in Children with Congenital Heart Disease Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2003; 7(2): 153 - 165. [Abstract] [PDF] |
||||
![]() |
C. Reichert Beyond halothane: an update on pediatric anesthesia pharmacology Can J Anesth, June 1, 2003; 50(90001): R10 - 10. [Full Text] |
||||
![]() |
P. J. Davis, R. L. Stiller, A. S. Wilson, F. X. McGowan, T. D. Egan, and K. T. Muir In Vitro Remifentanil Metabolism: The Effects of Whole Blood Constituents and Plasma Butyrylcholinesterase Anesth. Analg., November 1, 2002; 95(5): 1305 - 1307. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Rothstein Remifentanil for Neonates and Infants: Piano, Piano Con Calma Anesth. Analg., December 1, 2001; 93(6): 1370 - 1372. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|