Uptake pharmacokinetics of the Fentanyl Oralet�

in children scheduled for central venous accessremoval: implications for the timing of initiatingpainful procedures

MELISSA WHEELER M D*†, PATRICK K. BIRMINGHAM M D*†,

RICHARD M. DSIDA M D*†, ZHAO WANG M D†, CHARLES J.

COTE M D*†‡ AND MICHAEL J. AVRAM P h D†

*Department of Pediatric Anesthesiology, Children’s Memorial Hospital and †Departments ofAnesthesiology and ‡Pediatrics, North-Western University Medical School, Chicago, IL, USA

SummaryBackground: The Fentanyl Oralet� (Abbott Laboratories, Abbott Park,

IL, USA) is an oral transmucosal drug delivery system. We previously

examined pharmacokinetic parameters of children who had comple-

ted consumption of the Fentanyl Oralet�. The present study was

designed to clarify pharmacokinetic parameters during the con-

sumption phase to determine if there is an optimal administration

time before painful procedures.

Methods: Patients, aged 3–10 years, who were scheduled for elective

removal of central venous access devices under general anaesthesia,

received a Fentanyl Oralet� (fentanyl 10–15 lgÆkg)1). Plasma fentanyl

concentrations were measured by radioimmunoassay. Data from

blood samples obtained during and after consumption of the Fentanyl

Oralet� from 17 patients in the present study were combined with

data from our previous study to better characterize both the con-

sumption and postconsumption concentration versus time profiles.

Results: Estimated fentanyl bioavailability (mean ± SD) was low

(36.1 ± 0.4%), as were peak plasma concentrations

(1.03 ± 0.31 ngÆml)1), suggesting that many children swallowed a

large fraction of the dose. This led to a relatively late and variable peak

concentration time of 53 ± 40 min. In addition, because of the

apparently large degree of gastrointestinal absorption, concentration

versus time curves were wide and flat.

Conclusions: The wide and flat concentration versus time profile may

allow flexibility in the timing of a painful procedure following

Fentanyl Oralet� administration. However, the variability of the time

to peak concentration makes it difficult to suggest a minimum interval

between Fentanyl Oralet� consumption and the start of a painful

procedure.

Correspondence to: Melissa Wheeler, Department of Pediatric Anesthesiology, Children’s Memorial Hospital, Chicago, IL 60614, USA (e-mail:mwheeler@north-western.edu).

This study was presented in part at the Meeting of the American Society of Anesthesiologists, 1998, Orlando, FL, USA.

Paediatric Anaesthesia 2002 12: 594–599

594 � 2002 Blackwell Science Ltd

Keywords: anaesthesia; fentanyl; premedication; analgesia; fentanyl

Oralet�

Introduction

In children, painful procedures, such as lumbar

puncture and bone marrow aspiration, have often

been performed with little or no analgesia (1,2).

The adverse psychological and physiological con-

sequences for these children are beginning to be

appreciated (3). One reason for under treatment of

pain in children has been the inability to conveni-

ently administer opioid analgesics without addi-

tional discomfort for the child, such as that

associated with starting an intravenous (i.v.) line

or administering an intramuscular injection.

Nitrous oxide in oxygen has also been used to

provide analgesia but this requires specialized

equipment that may not be readily available.

Conventional oral analgesics also have limitations

as a premedicant for painful procedures in chil-

dren. These include antiplatelet properties, inad-

equate analgesia, delayed achievement of

therapeutic levels and limitations in dosage form,

such as pills that younger children are unable or

unwilling to swallow. The Fentanyl Oralet�

(Abbott Laboratories, Abbott Park, IL, USA) is an

oral transmucosal drug delivery system that pro-

vides a painless method of opioid administration

and is acceptable to most children. Limited data

on uptake pharmacokinetics are available for chil-

dren (4). The purpose of this study was to investi-

gate the early consumption pharmacokinetics of

the Fentanyl Oralet� to determine an optimal

administration time before painful procedures.

Methods

After Institutional Review Board approval, informed

consent was obtained from the parents of children,

aged 3–10 years, who were scheduled for elective

removal of central venous access devices under

general anaesthesia. Patients were ASA physical

status I or III and were excluded if they had a history

of clinically important renal, hepatic or cardiac

disease, or if they were allergic to fentanyl citrate.

No patients were receiving opioid therapy at the

time of study enrolment. Patients were given a

Fentanyl Oralet� in a dose range of 10–15 lgÆkg)1. A

research nurse observed each patient for side-effects

such as nausea, vomiting, pruritus and desaturation.

Oxygen saturation was monitored by pulse oximetry

throughout the study. Significant desaturation was

defined as a 5% decrease from baseline persisting

for more than 1 min (5). Parent and patient satisfac-

tion were recorded.

A 2-ml blood sample was drawn from the central

venous line at baseline and then every 5 min for

30 min during consumption of a Fentanyl Oralet�.

After consumption, additional blood samples were

drawn from either the central venous line or a

peripheral i.v. line placed after induction of anaes-

thesia. Using the start of Fentanyl Oralet� consump-

tion as time zero, these additional samples were

drawn at 40, 60, 80, 100, 120, 140, 160, 180, 200 and

220 min. Heparinized whole blood samples were

centrifuged and the plasma was separated and

frozen at )25 �C for subsequent analysis. Plasma

fentanyl concentrations were measured using a

radioimmunoassay kit (Research Diagnostics, Inc.,

Flanders, NJ, USA) (6) with average intra-assay and

interassay coefficients of variation of 6.0% and 6.9%,

respectively. Plasma samples with a concentration of

less than 0.1 ngÆml)1 were below the level of quan-

tification and were excluded from analysis.

Fentanyl plasma concentration versus time data

were analysed with a two-compartment open phar-

macokinetic model using SAAM II software (Version

1.12, SAAM Institute, Seattle, WA, USA) (7). To

better characterize maximal plasma concentration

(Cmax) and time to maximal plasma concentration

(tmax), we obtained frequent blood samples during

and shortly after consumption of the Fentanyl

Oralet�. Cmax and tmax were obtained from the data

collected in our current study. Data from the present

cohort were combined with those of a previous

study of both Fentanyl Oralet� and i.v. fentanyl

pharmacokinetics (8). In our previous study, we

did not collect data during the Fentanyl Oralet�

UPTAKE PHARMACOKINETICS OF THE FENTANYL ORALET 595

� 2002 Blackwell Science Ltd, Paediatric Anaesthesia, 12, 594–599

consumption phase because i.v. catheters were

inserted after induction of anaesthesia (8). Combi-

ning data allowed us to use a sequential i.v. and oral

dose modelling technique to better characterize

pharmacokinetic parameters. Data for characterizing

fentanyl disposition after rapid i.v. administration

were obtained from our previous study (8). Only

data from patients with consistent measurable

plasma fentanyl concentrations (n ¼ 14) were inclu-

ded in the current modelling because SAAM II uses

relative reciprocal data weighting, and therefore low

and incongruous data points significantly bias the

results. As the first step in our sequential modelling,

concentration versus time data from these 14

patients were fit to a two-compartment model with

bolus input (7).

As the second step in our sequential modelling,

volume of the central compartment (Vc), elimination

fractional transfer coefficient (k0,1) and intercompart-

mental fractional transfer coefficients (k1,2 and k2,1)

estimated by the two-compartment i.v. fentanyl

model were fixed. Then the plasma fentanyl con-

centration data from patients in both studies who

received the Fentanyl Oralet� (n ¼ 38) were ana-

lysed. An absorption compartment was added to

the two-compartment model, assuming first-order

absorption into the systemic circulation with an

absorption rate constant, ka, and fraction of the dose

absorbed, f.

Since fentanyl plasma samples were obtained

from two different studies, one with frequent samp-

ling during consumption and the other with samp-

ling only after consumption was completed, we used

multiple experiments modelling to calculate and

compare all associated model sample and data

elements to determine the optimal set of parameter

values for both the two-compartment i.v. pharmaco-

kinetic model and the ka and f of the Fentanyl

Oralet� absorption model (9). This is in contrast to

our previous study in which we used a naıve-pooled

pharmacokinetic modelling technique because there

was less variability in sampling times. In the current

analysis, data were weighted as the reciprocal of the

SD of the data estimate. Best fit was obtained by

assuming a fractional standard deviation of 0.5 for

i.v. data and a fractional SD of 0.1 for Fentanyl

Oralet� data. Both i.v. and oral doses were normal-

ized by body weight in order to reduce the variation

of the parameter estimation.

The a posteriori identifiability and appropriateness

of the choice of model order were verified using

fractional SDs and the Akaike information criterion,

respectively (9). The relationships between con-

sumption time and tmax or Cmax were sought using

standard least squares linear regression. P < 0.05

was considered statistically significant (10).

Results

Nineteen patients were enrolled in the present study.

One patient refused the Fentanyl Oralet�, and blood

samples from another patient were improperly

stored, leaving data from the 17 remaining patients

for analysis. Patients were aged 6.7 ± 4.9 years

(range 3.4–11.6 years), and weighed 21.5 ± 4.9 kg

(range 13.5–33 kg). The mean Fentanyl Oralet� dose

was 12.7 ± 1.2 lgÆkg)1 (range 10.6–14.8 lgÆkg)1). The

mean time to completion was 17 ± 7 min (range

6–33 min) and the time between completion and

entry to the operating room was 20 ± 18 min (range

2–66 min). Mean operative time was 50 ± 25 min

(range 16–120 min). A mean of 12 ± 1 blood samples

were collected from each patient. The data from these

17 patients were combined with 21 patients who

received the Fentanyl Oralet� in the previous study

for the second phase of modelling (7,8). The demo-

graphics of the 17 patients in the present study were

similar to those of our earlier study (5.6 ± 1.9 years;

23.9 ± 8.7 kg). The sampling and assay methodology

were identical in the two studies. The fentanyl assays

were all performed in the same laboratory and had

identical interassay and intra-assay coefficients of

variation. Individual plasma fentanyl concentrations

versus time profiles for all patients in both studies

who received the Fentanyl Oralet� are presented in

Figure 1.

The Akaike information criterion indicated that

the two-compartment model improved the statistical

fit compared with a one-compartment model but

was not improved by a three-compartment model.

All model parameter fractional SDs were 71% or less

and all but two were less than 50%. Pharmacokinetic

model parameters are summarized in Table 1. tmax

was extremely variable and only weakly related to

consumption time (tmax ¼ 3.2 · consumption

time ) 3.2, r2 ¼ 0.37, P ¼ 0.002). There was no rela-

tionship between consumption time and Cmax

(r2 ¼ 0.03).

596 M. WHEELER ET AL.

� 2002 Blackwell Science Ltd, Paediatric Anaesthesia, 12, 594–599

No patient in the present study required addi-

tional pain medication or had any preoperative or

postoperative desaturation. The only preoperative

side-effect observed in more than one patient was

pruritus (seven of 17 patients, 41.2%) and the only

postoperative side-effect observed in more than one

patient was vomiting (four of 17 patients, 23.5%)

(Table 2). All children who consumed the Fentanyl

Oralet� expressed satisfaction with the medication

modality. Both the patients and their parents stated

they would use it again for premedication for

general anaesthesia or before a painful procedure.

Discussion

This study investigated the early uptake pharmaco-

kinetics resulting from consumption of the Fentanyl

Oralet�. Our previous study was limited to samp-

ling times after consumption of the Fentanyl Oralet�

and after induction of anaesthesia since i.v. catheters

were not in place prior to premedication. The

current study specifically enrolled patients sched-

uled for removal of indwelling central venous

devices so as to clarify the early absorption phase

that we could not determine in our tonsillectomy

population. At the time of central venous device

removal, each patient had completed his course of

chemotherapy or antibiotics. We do not know if

chemotherapy or other treatment administered via

these devices would have affected fentanyl absorp-

tion, but we made an assumption that there was no

such effect.

The pharmacokinetic parameters in the present

study are similar to those reported for both children

and adults (8,11,12). Our ClI (31.0 ± 6.2 mlÆmin)1Ækg)1) is similar to that found in our previous

study (24.7 mlÆmin)1Ækg)1) and to the previously

reported two-compartmental paediatric value for i.v.

administered fentanyl (36 mlÆmin)1Ækg)1). Our ClE(15.2 ± 6.3 mlÆmin)1Ækg)1) is nearly identical to both

the paediatric (14.6 mlÆmin)1Ækg)1) and adult (19 mlÆ-min)1Ækg)1) values. Although our VDss

(4.87 ± 2.79 lÆkg)1) is larger than the paediatric value

we previously reported (2.44 lÆkg)1) and that of the

i.v. paediatric kinetic study (1.81 lÆkg)1), it is similar

to that reported for adults (3.51 lÆkg)1).

Our tmax of 53 ± 40 min is considerably longer

and more variable than that reported for adults

(23 ± 3 min) (13) and for a previous study of nine

children age 4–14 years (20 ± 2 min) (4). However,

our tmax for children is shorter and less variable than

that reported for adults given an oral fentanyl

solution (101 ± 48.8 min) (13). Because our tmax for

10.007.505.00

2.50

1.000.750.50

0.25

0.100 120 240 360

Time (min)

Pla

sma

fent

anyl

con

cent

ratio

n (n

g.m

l–1)

Data from present studyData from previous study

Figure 1Plasma fentanyl concentration versus time relationship for eachpatient receiving the Fentanyl Oralet�. Solid circles represent datafrom the present study (consumption phase and postconsumptionphase); open circles represent data from our previous study (8)(only postconsumption phase).

Table 1Oral transmucosal fentanyl citrate pharmacokinetic parameters

Parameter Estimate, mean ± SD (range)

tmax (min) 53 ± 40 (14–121)Cmax (ngÆml)1) 1.03 ± 0.31 (0.51–1.65)V1 (lÆkg)1) 1.3 ± 0.5V2 (lÆkg)1) 3.6 ± 2.5VDss (lÆkg)1) 4.9 ± 2.8ClI (mlÆmin)1Ækg)1) 31.0 ± 6.2ClE (mlÆmin)1Ækg)1) 15.2 ± 6.3t1 ⁄ 2b (h) 4.60f 0.361 ± 0.004ka (min)1) 0.019 ± 0.001

tmax, time to maximal concentration; Cmax, maximal concentration;V1, volume of central compartment; V2, volume of peripheralcompartment; VDss, volume of distribution at steady state; ClI,intercompartmental clearance; ClE, elimination clearance; t1 ⁄ 2b,elimination half-life; f, fentanyl bioavailability; ka, absorption rateconstant.

Table 2Side-effects of oral transmucosal fentanyl citrate (n ¼ 17)

Side-effect Preoperatively Postoperatively

Vomiting 1 (5.9%) 4 (23.5%)Nausea 1 (5.9%) 1 (5.9%)Pruritus 7 (41.2%) 1 (5.9%)Desaturation 0 0

UPTAKE PHARMACOKINETICS OF THE FENTANYL ORALET 597

� 2002 Blackwell Science Ltd, Paediatric Anaesthesia, 12, 594–599

the Fentanyl Oralet� in children is intermediate

between the adult Fentanyl Oralet� and oral solution

values, it suggests that both oral transmucosal and

gastrointestinal absorption of fentanyl is occurring.

The Cmax observed in the present study

(1.03 ± 0.31 ngÆml)1) is less than one-half the dose-

adjusted Cmax observed in adults administered the

Fentanyl Oralet� (2.37 ngÆml)1) (13), and two-thirds

of the dose-adjusted Cmax observed in children

administered the Fentanyl Oralet�(1.54 ngÆml)1)

(4). However, our Cmax was similar to the dose-

adjusted Cmax observed in adults administered an

oral fentanyl solution (1.35 ngÆml)1) (13). This fur-

ther suggests that, in our paediatric patients, there

was more gastrointestinal absorption than oral

transmucosal absorption with a resultant increase

in the amount of fentanyl experiencing first-pass

metabolism and therefore a decrease in its bioavail-

ability.

Fentanyl bioavailability in our paediatric patients

(0.36 ± 0.01) was similarly less than that observed in

adults administered the Fentanyl Oralet�

(0.50 ± 0.11) and nearly identical to that observed

in adults administered the oral fentanyl solution

(0.32 ± 0.10) (13).

The parameters of greatest interest for determin-

ing optimal timing of Fentanyl Oralet� administra-

tion for children who undergo a painful procedure

are tmax and Cmax. Plasma fentanyl concentration

versus time curves were relatively wide and flat

(Figure 1), suggesting ongoing uptake during and

following Fentanyl Oralet� consumption. These

curves are consistent with both oral transmucosal

and gastrointestinal fentanyl absorption. Wide and

flat plasma fentanyl concentration profiles suggest

that there would be a sustained opioid effect, thus

allowing some degree of flexibility in the timing of a

painful procedure following Fentanyl Oralet� con-

sumption. The Cmax observed in the present study is

well within the range of plasma fentanyl concentra-

tions providing analgesia without respiratory

depression in adults (0.6–2 ngÆml)1) (14), and is

unrelated to consumption time. However, because

tmax was quite variable and only weakly related to

consumption time, it is difficult to suggest a mini-

mum interval between Fentanyl Oralet� administra-

tion and the start of a painful procedure. Since our

previous study of tonsillectomy patients found an

11-fold variation in plasma fentanyl concentration at

the time of morphine rescue, a larger study may be

required to clarify the relationship between the

fentanyl levels produced by the Fentanyl Oralet�

and analgesia. This relationship may be determined

in part by the type of procedure.

As has been demonstrated for fentanyl-induced

glottic and chest wall rigidity, respiratory depres-

sion may depend in part on the rate of rise of opioid

concentrations, as well as peak plasma concentra-

tions (15). Clinically important, respiratory depres-

sion was not observed in the current study, our

previous study, or the study by Ginsberg and

colleagues, all of which used a Fentanyl Oralet�

dose of 10–15 lgÆkg)1 (8,16). The relatively slower

rate of rise and lower Cmax of fentanyl plasma

concentration with transmucosal delivery compared

with i.v. delivery may account for the lack of

observed desaturation events. However, respiratory

depression has been reported when the Fentanyl

Oralet� dose is between 15 and 20 lgÆkg)1 (17–20).

A study examining the use of the Fentanyl Oralet�

for procedural pain has suggested that the incidence

of nausea (12%) and vomiting (31%) may limit the

clinical usefulness of this medication (21). In our

study, the overall incidence of both side-effects was

comparable (nausea 12%, vomiting 29%). However,

only one patient had nausea and vomiting before

anaesthetic induction. In our previous study, we

theorized that our lack of adverse preoperative side-

effects may be due in part to the short interval

between Fentanyl Oralet� completion and anaes-

thetic induction (10 min) compared with other stud-

ies (30–90 min) (8,17–19). In our current study,

preoperative vomiting occurred in one of four

children who had an interval between Fentanyl

Oralet� completion and anaesthetic induction of

greater than 30 min. However, it is important to note

that our previous study found the incidence of

postoperative nausea and vomiting following Fen-

tanyl Oralet� administration to be no different from

that after i.v. fentanyl (8). Therefore, if a patient

requires an opioid, the route of administration does

not appear to affect the incidence of nausea and

vomiting.

The results of the present study suggest that

Fentanyl Oralet� may be useful in providing anal-

gesic fentanyl levels in children aged 3–10 years

who are undergoing painful procedures. How-

ever, since our data suggest a large degree of

598 M. WHEELER ET AL.

� 2002 Blackwell Science Ltd, Paediatric Anaesthesia, 12, 594–599

gastrointestinal absorption of fentanyl, the onset of

analgesia will be variable.

References1 Schechter NL, Blankson V, Pachter LM et al. The ouchless

place: no pain, children’s gain. Pediatrics 1997; 99: 890–894.2 Schechter NL. The undertreatment of pain in children: an

overview. Pediatr Clin North Am 1989; 36: 781–794.3 Weisman SJ, Bernstein B, Schechter NL. Consequences of

inadequate analgesia during painful procedures in children.Arch Pediatr Adolesc Med 1998; 152: 147–149.

4 Preston RA, Csontos ER, East KA et al. Plasma fentanylconcentrations after oral transmucosal fentanyl citrate: chil-dren versus adults. Anesthesiology 1993; 79: A370.

5 Cote CJ, Rolf N, Liu LM et al. A single-blind study of combinedpulse oximetry and capnography in children. Anesthesiology1991; 74: 980–987.

6 Michiels M, Hendriks R, Heykants J. A sensitive radioimmu-noassay for fentanyl. Plasma level in dogs and man. Eur J ClinPharmacol 1977; 12: 153–158.

7 SAAM Institute. SAAM II User Guide, Version 1.1. Seattle, WA:SAAM Institute, 1997.

8 Dsida RM, Wheeler M, Birmingham PK et al. Premedication ofpediatric tonsillectomy patients with oral transmucosal fenta-nyl citrate. Anesth Analg 1998; 86: 66–70.

9 Foster DM. Developing and testing integrated multicompart-ment models to describe a single-input multiple-output studyusing the SAAM II software system. Adv Exp Med Biol 1998;445: 59–78.

10 Akaike H. A new look at the statistical model identification.IEEE Trans Automat 1974; 19: 716–723.

11 Ginsberg B, Howell S, Glass PS et al. Pharmacokinetic model-driven infusion of fentanyl in children. Anesthesiology 1996; 85:1268–1275.

12 McClain DA, Hug CC Jr. Intravenous fentanyl kinetics. ClinPharmacol Ther 1980; 28: 106–114.

13 Streisand JB, Varvel JR, Stanski DR et al. Absorption andbioavailability of oral transmucosal fentanyl citrate. Anesthesi-ology 1991; 75: 223–229.

14 Peng PW, Sandler AN. A review of the use of fentanylanalgesia in the management of acute pain in adults. Anesthe-siology 1999; 90: 576–599.

15 Streisand JB, Bailey PL, LeMaire L et al. Fentanyl-inducedrigidity and unconsciousness in human volunteers. Incidence,duration and plasma concentrations. Anesthesiology 1993; 78:629–634.

16 Ginsberg B, Dear RB, Margolis JO et al. Oral transmucosalfentanyl citrate as an anaesthetic premedication when dosed toan opioid effect vs total opioid consumption. Paed Anaesth1998; 8: 413–418.

17 Friesen RH, Lockhart CH. Oral transmucosal fentanyl citratefor preanesthetic medication of pediatric day surgery patientswith and without droperidol as a prophylactic anti-emetic.Anesthesiology 1992; 76: 46–51.

18 Friesen RH, Carpenter E, Madigan CK et al. Oral transmucosalfentanyl citrate for preanaesthetic medication of paediatriccardiac surgery patients. Paed Anaesth 1995; 5: 29–33.

19 Epstein RH, Mendel HG, Witkowski TA et al. The safety andefficacy of oral transmucosal fentanyl citrate for preoperativesedation in young children. Anesth Analg 1996; 83: 1200–1205.

20 Streisand JB, Stanley TH, Hague B et al. Oral transmucosalfentanyl citrate premedication in children. Anesth Analg 1989;69: 28–34.

21 Schechter NL, Weisman SJ, Rosenblum M et al. The use of oraltransmucosal fentanyl citrate for painful procedures in chil-dren. Pediatrics 1995; 95: 335–339.

Accepted 15 May 2002

UPTAKE PHARMACOKINETICS OF THE FENTANYL ORALET 599

� 2002 Blackwell Science Ltd, Paediatric Anaesthesia, 12, 594–599