Anesth Analg 2002;94:44-49
© 2002 International Anesthesia Research Society
PEDIATRIC ANESTHESIA
The Pharmacokinetics of 
-Aminocaproic Acid in Children Undergoing Surgical Repair of Congenital Heart Defects
Douglas G. Ririe, MD,
Robert L. James, MS,
James J. O’Brien, MD,
Yonggu A. Lin, MS,
Judy Bennett, RN,
David Barclay, BS,
Michael H. Hines, MD, and
John F. Butterworth, MD
Departments of Anesthesiology and Cardiothoracic Surgery, Wake Forest University School of Medicine, Winston-Salem, North Carolina
Address correspondence and reprint requests to Dr. Ririe, Department of Anesthesiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157-1009. Address e-mail to dririe{at}wfubmc.edu
 |
Abstract
|
|---|
-Aminocaproic acid (ACA) is often administered to children undergoing cardiac surgery by using empiric dosing techniques. We hypothesized that children would have different pharmacokinetic variables and require a dosing scheme different from adults to maintain stable and effective serum ACA concentrations. Eight patients were enrolled in our study. ACA 50 mg/kg was administered three times IV: before, during, and after cardiopulmonary bypass (CPB). Nine serum samples were obtained. ACA plasma concentrations were measured by using high-performance liquid chromatography, and pharmacokinetic modeling was done by using NONMEM. The best fit was seen with a two-compartment model with volume of distribution (V1) adjusted for weight and CPB. Compared with published results in adults, modeling suggests that weight-adjusted V1 is larger in children than in adults before, during, and after CPB. Clearance from the central compartment (k10) was also greater in children than adults, and declined during CPB. Redistribution rates from the central compartment, k12 and k21, were greater in children and not affected by CPB. We modeled several different dosing regimens for ACA based on the larger V1, and higher redistribution and clearance variables. We conclude that, because of the developmental differences in pharmacokinetic variables of ACA, when compared with adult patients, a larger initial dose and faster infusion rate as well as an addi-tional dose on CPB are needed to maintain similar concentrations.
IMPLICATIONS: Pharmacokinetic modeling of -aminocaproic acid in children undergoing cardiac surgery suggests that there are developmental differences in pharmacokinetic variables. Based on these data, a dosing modification in children is suggested which may better maintain serum concentrations in children when compared with adults.
|
Introduction
|
|---|
-Aminocaproic acid (ACA), an inhibitor of fibrinolysis, is given to patients during cardiopulmonary bypass (CPB) to reduce postoperative bleeding and the need for transfusion. Empiric ACA dosing regimens seem to be safe, but have had varying efficacy (1,2). The variability in dosing, and therefore the variability in serum concentration, may be contributing to the varying efficacy. We previously reported results from adults whereby a smaller initial loading dose and a more rapid maintenance infusion maintained therapeutic serum concentrations better than most dosing regimens used (3). We also found that our weight-adjusted dosing technique in adults works equally well in older men and women (4).
As a result of these studies and the known pharmacokinetic differences in children, we hypothesized that children would have different pharmacokinetic variables from adults and might benefit from dosing regimens based on different modeling. We also hypothesized that, by collecting pharmacokinetic data in children, we could develop a dosing technique that would provide stable plasma ACA concentrations over time. This dosing technique would then permit an ACA concentration-versus-efficacy trial in children undergoing cardiac surgery in the future. Better dosing might also reduce the likelihood that children are exposed to a drug that is not providing any benefit because of inadequate serum concentrations.
|
Methods
|
|---|
After IRB approval, informed consent was obtained from parents to study 8 patients (aged 9 mo to 4 yr) undergoing repair of congenital heart defects with CPB (Table 1). All patients in the study received general anesthesia with an inhaled induction with sevoflurane. Patients older than 9 mo of age received midazolam 1 mg/kg oral premedication, whereas younger patients received no premedication. After peripheral IV access was established, a nondepolarizing neuromuscular blocker, either pancuronium or rocuronium, was given. Endotracheal intubation was performed and arterial and central venous access were established in all patients except the atrial septal defect patients who did not undergo central venous catheterization. General anesthesia was maintained with inhaled isoflurane and IV fentanyl (10–100 µg/kg).
CPB was conducted with hypothermia (target core temperature 28°C) by using a crystalloid priming volume of 650–850 mL. Blood was added to the pump prime if the calculated initial hematocrit on CPB based on the prime volume and the hematocrit measured in the operating room was estimated to be <16%. We used a Sarns (Ann Arbor, MI) Turbo 440 membrane oxygenator in every case. At the end of CPB, the contents of the venous reservoir were returned to the patient, or along with the remaining blood in the CPB apparatus, were transferred to the Cell Saver® (Hemonetics Inc., Braintree, MA), washed, and returned to the patient. Additional blood or blood products were administered after CPB as needed.
A set of timed arterial blood samples was obtained from each patient. Briefly, after heparin administration and before CPB, 50 mg/kg ACA was administered (over 5 min). Blood was drawn 2 and 5 min after the ACA infusion, just before CPB, and 3 min after initiation of CPB. ACA, 50 mg/kg, was then administered (as a bolus into the venous reservoir). Blood was obtained 3–5 and 10 min after the infusion, and just before separation from CPB. After CPB, ACA 50 mg/kg was infused (over 5 min) and blood was obtained 3–5 and 10 min after the infusion and just before leaving the operating room. The blood samples were immediately anticoagulated with EDTA and stored on ice. After each patient’s last sample had been obtained, the plasma was separated from blood cellular elements by centrifugation (1000g at 4°C), then the plasma was stored at -70°C pending analysis. Blood was analyzed by using high-performance liquid chromatography as described previously (3,5).
Statistical analyses were used to determine the best pharmacokinetic model to fit our data. Concentration-versus-time data from the present study were compared with the data from previous studies. The pharmacokinetic variables were determined (both measured from data and derived) and compared with the data from the adults.
Concentration-versus-time data were fit to compartmental models by using the nonlinear mixed-effects regression techniques of the NONMEM software package (NONMEM Project Group, University of California, San Francisco, CA). These pharmacokinetic models were fit modeling both fixed and random model variables by minimizing the -2 log likelihood objective function using NONMEM’s Laplacian estimation method. Constant CV (CCV), combined additive and CCV (Add + CV), and the power function were used to model intraindividual error structure (6).
The interpatient variability of the rate constants and volume of distribution (V1) was assumed to be lognormal in distribution and was modeled by NONMEM as follows:equation
where subscript i refers to the model variable and j indicates an individual patient, 
ij = variable estimate for individual patient j, i = estimate of population variable i, 
ij = random variable normally distributed with mean 0 and variance 
i that accounts for interpatient variability associated with the patient.
The ACA concentration-versus-time data were fit to the compartment models both with and without covariate adjustments for age and weight by using linear and quadratic functions. In addition, time-dependent covariates indicating before, during, and after CPB were tested for inclusion into the pharmacokinetic model. Model rate constants, k10, k12, k21, k20, etc., and the central compartment’s V1 were estimated directly by the NONMEM program. Variable subscripts refer to the model’s compartment number. Double subscripts refer to flow from one compartment to the next (e.g., k12 is the micro-rate constant describing drug movement from compartment 1 to compartment 2). Compartment 0 is outside the body. The model rate constants (min-1) are "fractional" clearance rates, i.e., the fraction of the drug in a compartment that is cleared by moving into the next compartment. "Standard" clearances (L/min), exponential coefficients, and half-lives were calculated by using standard formulae.
The Schwarz-Bayesian criteria were used to determine which models best fit the data. Models for which NONMEM was either unable to determine standard errors for, or those in which confidence intervals included zero, were excluded. Graphs showing model fits were used to confirm our choice of best model.
As a single measure of overall model performance, we evaluated the 75th, 90th, and 95th percentiles of the geometric performance error (GPE). The GPE for each sample is equal to the antilog (|log(observed) - log(predicted)|). Differences on the logarithmic scale become ratios on the arithmetic scale. Thus, a model with a 75th percentile of GPE = 1.52 means that 75% of the model-predicted concentrations are within a factor of 1/1.52 and 1.52 times (i.e., within 66% and 152%) of the observed concentration. We present the 50th and 95th percentiles of geometric performance error as well as the median relative predictive error.
All analyses were accomplished using either NONMEM or the SAS Program, version 8.0 (SAS Institute, Cary, NC) with P < 0.05 considered significant.
Dose regimens were simulated based on the pharmacokinetic variables for children from this study and from adults from previous studies (3,4).
|
Results
|
|---|
The ACA serum level increased rapidly after the initial bolus to a peak of 224 µg/mL (range 135–363) and then rapidly declined. The initiation of CPB produced a decrease of serum ACA level by 46% (range 28%–88%). After the bolus of ACA on CPB, the level peaked at 224 µg/mL (range 90–353 µg/mL). The serum levels once again declined after the initial bolus. No change in serum concentration was seen with termination of CPB. Serum levels of ACA increased again with a bolus after CPB to 221 µg/mL (range 170–267 µg/mL) and then declined. The best fit was seen using the two-compartment model with V1 adjusted for weight and CPB. Age was not a significant independent covariate after adjusting V1 for weight (weight being related to age). This model had a geometric performance error of 1.52 (75th percentile) and 1.98 (95th percentile), a median absolute predictive error of 22%. Data based on the observed and predicted values for each patient from the study are presented in Figure 1. Clearance from the central compartment (k10) was adjusted for CPB. Model variables are reported in Table 2 and derived variables in Table 3. Modeling suggests that V1 based on weight is larger in children (0.153 ± 0.026) than in adults (0.096 ± 0.006) before CPB. This difference between children and adults persists during CPB and after CPB (0.238 ± 0.02 for children and 0.144 ± 0.011 for adults). However, k10 in children (0.022 ± 0.002) was 2 times more than in the adults (0.011 ± 0.001) before and after CPB. During CPB, k10 in children (0.009 ± 0.003) was reduced to less than half of pre-CPB k10, whereas k10 in the adults (0.001 ± 0.001) was reduced to almost 0. The redistribution rates from the central compartment, k12 and k21, are both higher for children (0.035 ± 0.007 and 0.023 ± 0.005, respectively) than adults (0.012 ± 0.001 and 0.011 ± 0.001, respectively) and are not affected by CPB.
Based on the variables determined, we modeled several different dosing regimens for ACA. Although multiple infusion rates are needed for accurate maintenance of serum concentration, this is cumbersome and clinically not very practical. As a result, we have derived and simulated a more simple and reliable dosing regimen to maintain serum concentration at 2 times the serum levels to produce 95% efficacy (260 µg/mL level). This is 75 mg/kg over 10 min, 75 mg/kg on CPB, and 75 mg · kg-1 · h-1 infusion started at the end of the initial bolus, and maintained (Fig. 2). The bolus dose of 50 mg/kg and infusion rate of 25 mg · kg-1 · h-1 for adults reported previously using the fitted variables from children is also reported (Fig. 2B) and compared with the results from the adult model (Fig. 2B). Whereas the adult variables work well with this dosing (Fig. 2B), serum concentrations significantly less than therapeutic concentrations are present using the pediatric variables (Fig. 2B). We have also modeled dosing regimens from the only published data on efficacy of ACA in children (7) (data not shown) using the pediatric pharmacokinetic variables from our study. In smaller patients, the addition of ACA to the CPB prime maintained their peak serum concentrations better than the dosing scheme with no additional ACA in the CPB prime. Both regimens resulted in values significantly less than our presumed therapeutic target rate of 260 µg/mL.
View larger version (33K):
[in this window]
[in a new window]
|
Figure 2. A, Graphic simulation of our recommended dosing regimen of 75 mg/kg initial loading dose over 10 min, 75 mg/kg during cardiopulmonary bypass (CPB), and 75 mg · kg-1 · h-1 infusion started at the end of the first infusion/bolus. B, Graphic simulation using pediatric and adult variables and the recommended adult dosing regimen.
|
|
|
Discussion
|
|---|
Our data support the two-compartment model of the pharmacokinetics of ACA in children, as previously reported for adults (3). However, the notable differences in pharmacokinetic variables in children that we measured suggest that children require larger initial loading doses and larger maintenance infusions to maintain plasma ACA concentrations at or above the therapeutic threshold when compared with the initial loading dose and maintenance infusion requirements of adults.
Developmental differences in volumes of distribution are well recognized (8). These developmental differences result from the ways that extracellular, intracellular, plasma, and total body fluid compartmental volumes change as a function of development and age (9). The most frequent correlate of age and development in pharmacology is the weight of the patient in kilograms. Although the differences in fluid volumes do not always vary solely as a function of weight, weight is the most readily available predictor of development and organ maturity. ACA is a water-soluble drug and as such is more affected by the changes in body fluid compartments than more lipid-soluble substances. In our study, the larger V1 in children came as no surprise, given similar results for other water-soluble drugs such as ketorolac, lidocaine, and atracurium (10–12). Our results clearly suggest that children need a larger initial loading dose (on a per-kilogram basis) than adults.
Redistribution and elimination of drugs also change as a function of growth and development. The primary systems involved in metabolism and excretion of drugs are the hepatic and renal systems. The function of both of these systems is affected by age and CPB. Redistribution is affected through development in large part as a function of the changes in fluid compartments and the development of different organs (more adipose and muscle tissue per kilogram as age increases). Consistent with these differences, we measured increased k10, k12, and k21 in children. The increase in these variables in children results in a more rapid redistribution and elimination and suggests that children need a different maintenance dose than adults. Although we did not find age to be a significant independent covariate in our modeling after adjusting V1 for weight, a larger sample size may lead to age being a significant independent covariate even after adjustment of V1 for weight (weight being a function of age).
Although we do not know the "therapeutic" concentration of ACA, we used previous data and assumptions to calculate a target concentration. By using our mixed-effects model with adjustments for CPB, we calculated a dosing technique that targets an intraoperative ACA concentration of 260 µg/mL as was done previously in adult patients (3). We assumed that 130 µg/mL would completely inhibit fibrinolysis (13); however, based on our graphs of observed/predicted ACA concentrations, we anticipate that our model prediction could overestimate the actual concentration (in some patients) by as much as 50%. Thus, 260 µg/mL might be the more conservative target to minimize the likelihood of a patient having an unexpectedly low and potentially ineffective ACA concentration. We assume that a certain concentration must be exceeded in all patients for consistent efficacy, but we recognize that this is an untested hypothesis. We also recognize that there are no data to suggest a maximal concentration above which thrombotic and other complications of ACA may be more likely. Nevertheless, we believe that it is unwise to expose patients to larger than necessary ACA concentrations.
Simulations based on our data suggest that multiple infusion rates are necessary to maintain nearly constant plasma concentrations over time in children. Multiple infusion rates are cumbersome and impractical clinically. Therefore, based on simulation from these data, in clinical use, a simple and more practical approach would be to use an initial loading dose of 75 mg/kg over 10 min and a maintenance infusion rate of 75 mg · kg-1 · h-1 with 75 mg/kg placed in the pump to maintain serum concentrations significantly more than the therapeutic level (assumed to be 130 µg/mL) in >95% of patients.
At the initiation of CPB, the volume of distribution increases in both children and adults. Because the volume of the CPB circuit is a larger volume in children relative to the blood volume when compared with adults, the change in plasma concentrations of ACA is much larger in children and not compensated by the maintenance infusion. Furthermore, the k10 is reduced to nearly nil in the adults, but maintained at a reduced value in children, creating even more discrepancy between the adults’ and children’s plasma concentrations during CPB. Therefore, in children, the addition of an initial loading dose into the CPB circuit reduces the effect of the initiation of CPB on the serum concentration of ACA during and after CPB. This is the rationale for the addition of 75 mg/kg ACA into the CPB circuit in the simulation in children, whereas in the adult model, no such additional dosing is needed when commencing CPB.
It is important to realize that our modeling was done for patients on CPB with a core target temperature of 28°C. However, many pediatric cardiac operations are done with deep hypothermia. This study provides no information on pharmacokinetic changes on CPB and afterward in the subgroup of pediatric patients undergoing deep hypothermia with or without circulatory arrest.
The use of ACA in children for cardiac surgery is not a new concept. In 1963, in the first published work regarding use of ACA in children, an initial loading dose of between 64 and 180 mg/kg was used (14). Interestingly, in the only study of clinical efficacy of ACA in children during open heart surgery and CPB, an initial loading dose of 75 mg/kg was used (7). However, based on our simulation data, the infusion rate these authors used (15 mg · kg-1 · h-1) was inadequate. These authors added 510 mg total of ACA in the CPB prime for each unit of blood that was included. Unfortunately, this investigation does not report which patients received blood in the CPB prime, or the weight of patients who had excessive bleeding. In smaller patients, or in patients with lower preoperative hematocrit, blood would more likely be added to the CPB prime, and with it, an extra dose of ACA would have been administered, increasing the likelihood of therapeutic plasma ACA concentrations. We modeled this dosing regimen for a 20-kg child. When the dose of ACA was given in the CPB prime with the blood, a therapeutic serum level was achieved during CPB, but not maintained. This is related to the effects of CPB on V1 and the need for another dose during CPB as mentioned. However, the absence of an adequate infusion failed to maintain therapeutic serum concentrations.
Our measured pharmacokinetic variables in children are very different from those measured in adults, suggesting the need for dosing modification in children. Further study to validate the dosing regimen in children predicted from simulation using our pharmacokinetic variables from this study will be needed. Studies in children undergoing deep hypothermia are also warranted to determine how further decreases in temperature change pharmacokinetic variables. Furthermore, accurate evaluation of ACA concentration versus efficacy in children undergoing cardiac surgery will allow determination of the minimal effective serum concentration. Further studies will also be needed to determine at what age or weight the child assumes a pharmacokinetic profile seen in adult patients. Future studies should lead to safer and more consistently effective dosing in children undergoing heart surgery with CPB.
|
References
|
|---|
-
Horrow JC, Fitch JCK. Management of coagulopathy associated with cardiopulmonary bypass. In: Gravlee GP, Davis RF, Kurusz M, Utley JR, eds. Cardiopulmonary bypass: principles and practice. 2nd ed. Baltimore: Williams & Wilkins, 2000: 506–33.
-
Carr ME Jr. Control of perioperative bleeding: pharmacologic agents. In: Wechsler AS, ed. Pharmacologic management of perioperative bleeding. Southampton, NY: CME Network Publishing, 1996: 28–34.
-
Butterworth J, James RL, Lin Y, et al. Pharmacokinetics of epsilon-aminocaproic acid in patients undergoing aortocoronary bypass surgery. Anesthesiology 1999; 90: 1624–35.[ISI][Medline]
-
Butterworth J, James RL, Lin Y, et al. No difference in epsilon-aminocaproic acid concentrations or pharmacokinetics between males and females [abstract]. Anesth Analg 2000; 90: SCA65.
-
Davey JF, Ersser RS. Amino acid analysis of physiological fluids by high-performance liquid chromatography with phenylisothiocyanate derivatization and comparison with ion-exchange chromatography. J Chromatogr 1990; 528: 9–23.[ISI][Medline]
-
Boeckmann AJ, Sheiner LB, Beal SL. NONMEM users guide. Part V. Introductory guide. For the NONMEM Project Group. San Francisco: Regents of the University of California, 1994.
-
McClure PD, Isaak J. The use of epsilon-aminocaproic acid to reduce bleeding during cardiac bypass in children with congenital heart disease. Anesthesiology 1974; 40: 604–8.[Medline]
-
Gregory GA. Pharmacology. In: Gregory GA, ed. Pediatric anesthesia. 3rd ed. New York: Churchill-Livingstone, 1994: 13–45.
-
Stoelting RK. Body fluids. In: Stoelting RK, ed. Pharmacology and physiology in anesthetic practice. Philadelphia: JB Lippincott, 1987: 577–82.
-
Fisher DM, Canfell PC, Spellman MJ, Miller RD. Pharmacokinetics and pharmacodynamics of atracurium in infants and children. Anesthesiology 1990; 73: 33–7.[ISI][Medline]
-
Olkkola KT, Maunuksela EL. The pharmacokinetics of postoperative intravenous ketorolac tromethamine in children. Br J Clin Pharmacol 1991; 31: 182–4.[ISI][Medline]
-
Finholt DA, Stirt JA, DiFazio CA, Moscicki JC. Lidocaine pharmacokinetics in children during general anesthesia. Anesth Analg 1986; 65: 279–82.[Abstract/Free Full Text]
-
Frederiksen MC, Bowsher DJ, Ruo TI, et al. Kinetics of epsilon-aminocaproic acid distribution, elimination, and antifibrinolytic effects in normal subjects. Clin Pharmacol Ther 1984; 35: 387–93.[Medline]
-
Weiss M, Yver J, Samama M, Dubost Ch. Effets dissociés d’un nouvel antifibrinolytique (acide epsilon-amino-caproïque) injecté in vivo, sur les épreuves in vitro de fibrinolyse. Presse Med 1963; 71: 1879–80.[Medline]
Accepted for publication September 7, 2001.
This article has been cited by other articles:
|
|
|
|
 
J. Butterworth, Y. A. Lin, R. C. Prielipp, J. Bennett, J. W. Hammon, and R. L. James
Rapid disappearance of protamine in adults undergoing cardiac operation with cardiopulmonary bypass
Ann. Thorac. Surg.,
November 1, 2002;
74(5):
1589 - 1595.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
