Plasma protein binding is a reversible interaction between a drug molecule and a protein [1-5]. This interaction may be affected by the concentration of the drug and protein available for binding. The therapeutic and toxic effects of a drug are often related to the amount of free drug in the blood stream.
Although opioid anesthesia provides hemodynamic stability, adequate intraoperative anesthesia, and postoperative analgesia in seriously ill neonates, there is scant information available on the pharmacology of opioids in newborn infants [6-8]. Alfentanil and fentanyl are two opioids that are highly protein bound to alpha1-acid glycoprotein (AAG); however, their binding properties in term and preterm neonates have not been elucidated . Small changes in protein binding of highly bound drugs can significantly affect their pharmacodynamics and kinetics. Because clinicians often consider fentanyl and its congeners to be interchangeable at appropriate doses, and because alfentanil and fentanyl are often used to provide anesthesia/analgesia for neonates, we undertook this study to determine the effects of gestational age (GA) on fentanyl and alfentanil protein binding.
After institutional review board approval, umbilical cord blood samples (1.0 mL) were obtained from 72 newborns at the time of their delivery. The neonates were divided into two groups based on GA. Neonates were considered to be preterm at less than 37 wk gestation and term at GAs of 37 wks or greater. Blood samples were collected in a heparinized syringe and immediately centrifuged. The plasma layer was separated and stored at -70 degrees C.
In separate equilibrium dialysis experiments, fentanyl (125 ng/mL) or alfentanil (125 ng/mL) was added to 200 micro L plasma sample. These concentrations were chosen so that the concentrations of the two drugs were equal, and the limit of quantitation for the analysis of alfentanil was achieved. Because of the concern that fentanyl binding may be altered at 125 ng/mL, we determined in separate studies that fentanyl binding was independent of concentration in the range of 10-500 ng/mL (71% +/- 0.7%). The pH of the sample was adjusted to the range of 7.4-7.5 using 0.1 M monobasic potassium phosphate buffer. Fentanyl (125 ng/mL) or alfentanil (125 ng/mL) was added to 200 micro L 0.1 M monobasic phosphate buffer, pH 7.4, as a control. Equilibrium dialysis was composed of a split cell containing plasma in one half of the cell separated by a membrane from a buffer medium in the other half of the cell. The dialysis cells were rotated at 20 rpm in a thermostat bath at 37 degrees C for 4 h. At equilibrium (4 h), the drug in the buffer cell half equaled the free drug concentration (Cf) and the drug in the plasma cell half equaled the sum of the concentration (C) of both the free and bound drug (Cf + Cb = C). Therefore, the percentage of free drug = Cf/C x 100, and the percentage of bound drug = C - Cf/C x 100.
Alfentanil and fentanyl concentrations in each half cell were determined with a specific radioimmunoassay (Research Diagnostics, Flanders, NJ). The limit of quantitation of the alfentanil and fentanyl assays was 0.1 ng/mL, and the coefficient of variation was less than 6%. The loss of alfentanil and fentanyl to nonspecific cell binding was minimal (less than 4%).
AAG and albumin were measured with a radial immunodiffusion assay procedure. The limit of quantitation of the NanoRID[TM] AAG assay (The Binding Site, Birmingham, England) was 1.69 mg/L and had a coefficient of variance of 3.7%. The limit of quantitation of the NOR-Partigen[TM] albumin assay (Behring Diagnostics, Somerville, NJ) was 800 mg/dL and had a coefficient of variation of 8.2%.
The differences between preterm and term neonates were analyzed using unpaired t-tests. The correlation of the binding ratio (percentage of bound drug/percentage of free drug) with protein concentrations (AAG and albumin) and age was determined by multiple regression analysis. Although age was a continuous variable in the regression, we also compartmentalized age into two defined groups (preterm and term). Significance was considered P < 0.05.
There were significant differences in age, free alfentanil, free fentanyl, and AAG concentration between the term and preterm neonates (Table 1). There was no significant difference in albumin concentration between the two groups.
Using multiple variate regression analysis, a positive correlation was found between the alfentanil binding ratio and GA, AAG, and albumin (r2 = 0.5566, P = 0.0000), and a weak correlation was found between the fentanyl binding ratio and GA, AAG, and albumin (r2 = -0.1848, P < 0.000012). The regression coefficients, 95% confidence intervals, and P-values for alfentanil and fentanyl binding versus GA, AAG, and albumin can be found in Table 2 and Table 3. The regression curves are shown in Figure 1 and Figure 2.
In this study, we noted that GA influences the unbound fraction of the opioids fentanyl and alfentanil in nonpredictable ways. Because the free, unbound portion of a drug is usually the pharmacologically active moiety, these opioids may not be interchangeable at the appropriate clinical doses used during anesthesia for infants in the newborn period. In addition to the plasma protein binding of alfentanil and fentanyl, we also measured albumin and AAG concentrations. The levels of AAG in term infants were in agreement with those previously reported [9-11]. An increase in AAG was observed with an increase in GA, but there was no change in albumin concentration with GA. We observed a greater plasma concentration of free alfentanil in preterm than term neonates. This increase in the preterm infants' free drug fraction may have been a function of the age-related changes of AAG as opposed to the albumin concentration.
Age-related AAG changes have also been associated with sufentanil protein binding. Meistelman et al.  and others have noted that sufentanil protein binding is inversely related to AAG concentrations. Although term neonates had a significantly higher free drug fraction than infants, small children, and adults, this study did not evaluate sufentanil binding in premature neonates . Lerman et al. also noted age-related effects of AAG with serum lidocaine binding. Although similar relationships of AAG with age and protein binding were noted, this study's neonatal population involved cord blood samples from only five preterm (30-36 weeks' gestation) neonates and no term infants. Thus, the differences due to GA could not be ascertained .
The age-related changes in fentanyl binding were surprising in that they were different from the binding results observed with alfentanil. We noted a greater percentage of free drug in term infants than in preterm infants. The regression results show a weak but negative correlation between the fentanyl binding ratio and GA (r (2) = -0.1066). There was no correlation of fentanyl binding with AAG or albumin concentrations. One of the possible criticisms of our study was that we determined plasma protein binding at 125 ng/mL for both fentanyl and alfentanil. Although this is a clinically relevant concentration for alfentanil, this suprapharmacologic level for fentanyl may have altered (decreased) fentanyl's in vitro binding. This appears to be unlikely, because fentanyl protein binding has been reported to be constant over a 100-fold difference in plasma concentration [5,13]. In addition, we noted no changes in fentanyl protein binding when plasma fentanyl concentrations ranged from 10 to 500 ng/mL. Alfentanil protein binding is constant over plasma concentrations from 0.1 to 1000 ng/mL .
The increased free fentanyl concentration with GA, coupled with the two-fold increase in AAG with age, suggests that there may be a factor in the plasma that inhibits binding. The effect of this factor may show a stronger correlation to the fentanyl binding ratio than to GA. It has been documented that the newborn has decreased plasma protein binding for many drugs. Although a reduced amount of plasma proteins, i.e., albumin and AAG, is often cited as a factor in decreasing drug binding, inhibitors present in the plasma, such as 2-hydroxybenzoylglycine, have been shown to affect plasma protein binding in newborns .
Another factor that may have influenced the in vitro drug binding in this study was plasma pH . The significance of the pH effect is dependent on the negative logarithm of acid ionization constant (pKa) of the drug. Fentanyl (pKa = 8.43) binding is significantly affected by small deviations from normal physiological pH; however, alfentanil (pKa = 6.50) binding is not adversely affected by these small deviations. Although the protein binding of basic drugs is dependent on the pH of the plasma, a pH effect on binding was unlikely in view of the fact that pH was maintained throughout the experiment.
AAG is an acute phase reactant, and its plasma concentrations increase dramatically during the first year of life and even within the first weeks of life . Thus, our data should be applied to newly born neonates and not extrapolated to infants with similar postconceptual age. For clinicians who care for newborn infants, opioids are often used interchangeably at appropriate doses, provided the potency and the pharmacokinetics profile of the drugs are known. This study suggests that for newly born neonates, the drugs are not interchangeable at the GAs studied because of the effect that GA has on protein binding.
1. Suh B, Wadsworth S, Lichtenwalner D. Demonstration of 2-hydroxybenzoylglycine as a drug binding inhibitor in newborn infants. J Clin Invest 1987;80:1125-31.
2. Bowdle T, Horiya A, Kharasch E. The pharmacologic basis of anesthesiology. New York: Churchill Livingstone, 1984.
3. Herngren L, Ehrnebo M, Boreus L. Drug binding to plasma proteins during human pregnancy and in the perinatal period: studies on cloxacillin and alprenolol. Dev Pharmacol Ther 1983;6:110-24.
4. Paxton J. Alpha1-acid
glycoprotein and binding of basic drugs. Methods Find Exp Clin Pharmacol 1983;5:635-48.
5. Meuldermans W, Hurkmans R, Heykants J. Plasma protein binding and distribution of fentanyl, sufentanil, alfentanil and lofentanil in blood. Arch Int Pharmacodyn 1982;257:4-19.
6. Davis PJ, Killian A, Stiller RL, et al. Pharmacokinetics of alfentanil in newborn premature infants and older children. Dev Pharmacol Ther 1989;13:21-7.
7. Marlow N, Weindling AM, Cooke RWI. Hazards of analgesia for newborn infants [letter]. Arch Dis Child 1988;63:1293.
8. Killian A, Davis PJ, Stiller RL, et al. Influence of gestational age on pharmacokinetics of alfentanil in neonates. Dev Pharmacol Ther 1990;15:82-5.
9. Meuldermans W, Woestenborghs R, Noorduin H, et al. Protein binding of the analgesics alfentanil and sufentanil in maternal and neonatal plasma. Eur J Clin Pharmacol 1986;30:217-9.
10. Sann L, Bienvenu J, Lahet C, et al. Serum orosomucoid concentration in newborn infants. Eur J Pediatr 1981;136:181-5.
11. Meistelman C, Benhamou D, Barre J, et al. Effects of age on plasma protein binding of sufentanil. Anesthesiology 1990;72:470-3.
12. Lerman J, Strong HA, LeDez KM, et al. Effects of age on the serum concentration of alpha1-acid
glycoprotein and the binding of lidocaine in pediatric patients. Clin Pharmacol Ther 1989;46:219-25.
© 1997 International Anesthesia Research Society
13. Mather LE. Clinical pharmacokinetics of fentanyl and its newer derivatives. Clin Pharmacokin 1983;8:422-46.