Double-blind, placebo-controlled analgesic study of ibuprofen or

rofecoxib in combination with paracetamol for tonsillectomy in

children
A. E. Pickering, H. S. Bridge, J. Nolan and P. A. Stoddart
*
Sir Humphry Davy Department of Anaesthesia, Bristol Royal Hospital for Children, Bristol BS2 8BJ, UK
*Corresponding author
Background. The analgesics used for paediatric tonsillectomy may be associated with side-
effects such as sedation, respiratory depression and vomiting (opioids) or increased bleeding
[non-steroidal anti-inammatory drugs (NSAIDs)]. In our institution, we employ a combination
of paracetamol, NSAID and opioid, although there is no published evidence of analgesic benet
from adding NSAIDs to paracetamol in children.
Methods. This randomized, double-blinded clinical study examined the analgesic effectiveness
of combining paracetamol (20 mg kg
1
) with rofecoxib (0.625 mg kg
1
), ibuprofen (5 mg kg
1
)
or placebo as premedication for (adeno)tonsillectomy (n=98) in children aged 315 yr.
Intravenous fentanyl 12 mg kg
1
was given intraoperatively. Regular oral paracetamol
(15 mg kg
1
, 4 hourly) was given after operation and could be supplemented on request from
the child with oral ibuprofen 5 mg kg
1
or oral codeine 1 mg kg
1
. The primary outcome
variable was need for early supplementary analgesia (within 2 h after surgery).
Results. The addition of ibuprofen to paracetamol reduced the need for early analgesia from
72% to 38% of children (difference 34%; 95% condence interval 464%). The addition of
rofecoxib to paracetamol did not signicantly alter the need for early analgesia (68 vs 72%).
Pain scores were higher in those children who required early analgesia. There were no differ-
ences between the groups in operative blood loss or complications, total 24-h analgesic con-
sumption, pain scores at 4 and 8 h, vomiting or antiemetic use.
Conclusion. This study provides evidence to support the combination of ibuprofen (but not
rofecoxib) with paracetamol for perioperative analgesia in children.
Br J Anaesth 2002; 88: 727
Keywords: analgesics non-opioid, ibuprofen; analgesics non-opioid, paracetamol; analgesics
non-opioid, rofecoxib; antagonists, cyclo-oxygenase type 2 antagonist; surgery,
otolaryngological; surgery, paediatric
Accepted for publication: August 31, 2001
Tonsillectomy is one of the most common surgical pro-
cedures in children, with an estimated yearly rate in the UK
of 2.3/1000 for the population aged under 12 yr. Increasing
attention is being paid to the control of perioperative pain in
children with the development of validated pain scales and
more vigorous employment of analgesic techniques.
1
However, the provision of adequate analgesia after
tonsillectomy presents the clinician with difculties as this
is a painful procedure and it may be associated with
signicant bleeding into the airway. Opioids are frequently
used and provide effective analgesia for tonsillectomy
2
but
are associated with a high incidence of nausea and
vomiting.
3
In addition, there are concerns about opioid-
induced respiratory depression and sedation, which may
make recovery hazardous after pharyngeal surgery,
4 5
particularly in children with a history of obstructive sleep
apnoea. Non-steroidal anti-inammatory drugs (NSAIDs)
have a similar analgesic effect to opioids for tonsillectomy
5
and are opioid-sparing in the recovery period.
4
However,
the perioperative use of NSAIDs has been limited because
of concerns over increased postoperative bleeding, which
has been demonstrated with ketorolac,
6 7
although appar-
ently not for other NSAIDs.
8
Paracetamol has been shown to
be an effective and relatively safe analgesic in children
9 10
British Journal of Anaesthesia 88 (1): 727 (2002)
The Board of Management and Trustees of the British Journal of Anaesthesia 2002
and is often used to provide a background of analgesia after
surgery, but appears to be insufciently potent if used alone
for tonsillectomy pain.
11
The current practice in our unit is to use a balanced
approach
12
with a combination of regular paracetamol and
NSAIDs plus opioid as needed to provide perioperative
analgesia for tonsillectomy.
13
However, there is no pub-
lished evidence to support the use of NSAIDs with
paracetamol for perioperative analgesia in children. A
recent study of children after appendectomy failed to show
any additional morphine-sparing effect from combining
paracetamol with diclofenac.
14
This contrasts with an adult
study, which showed the combination of paracetamol and
diclofenac to reduce postoperative morphine consumption
compared with either drug alone.
15
The recently introduced cyclo-oxygenase type 2 (COX-2)
inhibitors, such as rofecoxib,
16
may have advantages for
analgesia in paediatric tonsillectomy. They have analgesic
power equivalent to that of NSAIDs for dental extraction
pain.
17 18
Their COX-2 selectivity produces minimal effect
on platelet aggregation. In addition, rofecoxib has a syrup
formulation, excellent oral bioavailability and a half-life of
action of 17 h, allowing once-daily dosing. These properties
potentially make it a good candidate for perioperative use in
paediatrics to help reduce `tears at bedtime'.
19
Currently,
there are no pharmacokinetic or pharmacodynamic data on
the use of COX-2 antagonists in children. Rofecoxib 50 mg
has been shown to be equianalgesic to ibuprofen 400 mg in
the adult dental pain model.
17
We have used this measure of
the potency of rofecoxib to estimate an equianalgesic
paediatric dose (0.625 mg kg
1
) as being one-eighth of the
paediatric ibuprofen dose (5 mg kg
1
).
The aim of our randomized, double-blinded, placebo-
controlled pragmatic clinical study was to examine whether
combining a NSAID (ibuprofen 5 mg kg
1
) or COX-2
antagonist (rofecoxib 0.625 mg kg
1
) with paracetamol
(20 mg kg
1
) before paediatric tonsillectomy would confer
analgesic benets without increasing intraoperative blood
loss.
Methods
The study was approved by the hospital ethics committee. A
Clinical Trials Exemption certicate was obtained from the
Medicines Control Agency of the UK to allow the use of
rofecoxib as an analgesic in children. After written informed
parental consent and verbal child assent had been obtained,
103 children, ASA I or II, aged 315 yr, presenting for
elective tonsillectomy or adenotonsillectomy, were entered
in the trial. Using a computer-generated random number
table, a hospital pharmacist allocated the children randomly
to three groups: placebo, rofecoxib or ibuprofen. Both the
patients and the investigators were blinded to the study
group assignment. Exclusion criteria were sensitivity to
NSAIDs, bleeding diathesis, renal or hepatic impairment
and severe asthma (requiring recent hospital admission or
oral corticosteroid therapy). On enrolment in the study, the
children were familiarized with the Oucher visual analogue
pain scale, which has been validated in children
20
and used
previously to score tonsillectomy pain.
21
The children were premedicated 1 h before surgery with
paracetamol syrup 20 mg kg
1
and either placebo, ibuprofen
5 mg kg
1
or rofecoxib 0.625 mg kg
1
(syrups 0.25 ml kg
1
to a maximum of 10 ml). Therefore, the maximum dose was
ibuprofen 200 mg or rofecoxib 25 mg. Induction of
anaesthesia was with propofol or sevourane and the
airway was secured with a reinforced laryngeal mask
(53% of children) or endotracheal tube (47%). Atracurium
0.5 mg kg
1
was used to facilitate intubation. Anaesthesia
was maintained with isourane in 33% oxygen/67%
nitrous oxide and intraoperative analgesia was provided
with fentanyl 12 mg kg
1
. No NSAIDs, paracetamol or
prophylactic antiemetic were given in theatre. Intra-
operative blood loss was assessed by weighing swabs
(assuming 1 g=1 ml blood) and by measuring losses to
calibrated suction bottles.
A standardized postoperative analgesic regimen was
prescribed of regular paracetamol (15 mg kg
1
, 4 hourly)
with oral ibuprofen (5 mg kg
1
) or oral codeine (1 mg kg
1
)
to be administered at the child's request by the ward nursing
staff. Pain was scored whenever supplementary analgesia
was administered and also 4 and 8 h after premedication,
when the the rst two postoperative doses of paracetamol
were given. Children were considered to have required early
supplementary analgesia if they received either ibuprofen or
codeine within 2 h after surgery. Children were discharged
home the day after surgery.
The primary outcome variable was the need for early
postoperative analgesia. Secondary outcome variables were
intraoperative blood loss, time to rst postoperative anal-
gesia, Oucher pain scores, analgesic consumption in the rst
24 h, vomiting, antiemetic use and incidence of primary and
secondary haemorrhage. The data were collected by the
anaesthetist, theatre and ward nurses and subsequently
validated by review of the clinical notes.
Differences among groups were compared using the
KruskalWallis test for non-parametric data, one-way
analysis of variance for parametric data and the c
2
test or
Fisher's exact test for nominal data. P<0.05 was considered
signicant. A sample size calculation based on audit data
from our hospital showed that 60% of children pre-
medicated with paracetamol and receiving rectal diclofenac
intraoperatively required early postoperative analgesia.
Therefore, to show a 50% reduction in the need for early
analgesia in the rofecoxib group, we estimated that 42
patients were required per group (power of 80% and
P<0.05). To minimize the size of the placebo group, a
blinded interim analysis was planned by the statistician at
the halfway point. If this showed a greater need for early
postoperative analgesia in the placebo group than the
ibuprofen group, then recruitment to the placebo group
would cease.
Analgesia for paediatric tonsillectomy
73
Results
Of the 103 children recruited, ve were excluded because of
protocol violations (two were too young and three received
diclofenac in theatre). The patient and surgical character-
istics were comparable across the groups (Table 1). There
were no differences in the time from premedication to the
start of surgery [78 (29) min; pooled mean (SD)], operation
type (tonsillectomy or adenotonsillectomy), duration of
surgery or dose of intraoperative fentanyl (Table 1).
The interim blinded analysis showed that more children
in the placebo group received early postoperative analgesia
than in the ibuprofen group (72 vs 38%; difference 34%,
95% condence interval 464%). Therefore recruitment to
the placebo group was stopped at 18 children. Signicantly
more children required early analgesia (within 2 h after
surgery) in the rofecoxib group compared with the
ibuprofen group (68 vs 43%; difference 25%, 95% con-
dence interval 446%) (Table 2). The pain scores at the
time of administration of supplementary analgesia were
signicantly higher in those children requesting early
analgesia (median 60 vs 40, P<0.0001). By the time of the
rst postoperative paracetamol dose, there was no differ-
ence in pain scores between the groups (Table 2). There was
no difference between the groups in total analgesic
consumption in the rst 24 h, in the incidence of vomiting
or the need for antiemetic (Table 2).
There was greater surgical blood loss in children having
adenotonsillectomy than in those having tonsillectomy
[median (interquartile range) 2.6 (1.63.9) vs 0.71
(0.311.4); P<0.0001]. There was no signicant difference
between the treatment groups in blood loss for either
tonsillectomy or adenotonsillectomy (Fig. 1), nor was there
any difference in the incidence of primary (bleeding
requiring intervention within 24 h) or secondary (bleeding
requiring readmission) haemorrhage (Table 2). Two of the
children required a second operation to achieve adequate
haemostasis (one each in the placebo and ibuprofen groups).
Both rofecoxib and ibuprofen were well tolerated by the
children, with no refusal of medication and no episodes of
bronchospasm, pruritis or dizziness. One child in the
ibuprofen group developed a mild erythematous, non-itchy
rash after surgery, which resolved within 4 h without
treatment.
Discussion
Ibuprofen in combination with paracetamol as premedica-
tion reduced the need for early supplementary analgesia by
almost 50% after tonsillectomy. The clinical importance of
this nding is indicated by the higher pain scores at the time
of supplementary analgesia in the children requiring early
rescue. This equates to fewer distressed children on the ward
Table 1 Patient characteristics and intraoperative details. Data are median (interquartile range) or mean (SD). There were no signicant differences between
the groups in any of the variables. T=tonsillectomy; T&A=tonsillectomy and adenoidectomy
Drug group
Placebo (n=18) Rofecoxib (n=40) Ibuprofen (n=40)
Excluded (n) 1 2 2
Age (yr) 7.0 (59) 7.5 (511) 7.0 (511)
Weight (kg) 25 (2031) 29 (2149) 28 (2142)
Sex (male:female) 7:11 16:24 21:19
Asthmatic (%) 4 (22%) 6 (15%) 2 (5%)
Time from premedication to surgery (min) 81 (33) 73 (28) 82 (28)
Operation type (T:T&A) 11:7 22:18 22:18
Operation duration (min) 20.1 (8.5) 21.0 (10.0) 21.2 (9.0)
Fentanyl dose (mg kg
1
) 1.46 (0.6) 1.34 (0.4) 1.37 (0.4)
Table 2 Primary and secondary outcomes. Data are percentage (n) or median (interquartile range). `Primary haemorrhage' indicates bleeding requiring
intervention within 24 h; `Secondary haemorrhage' indicates delayed bleeding requiring readmission
Drug group P
Placebo Rofecoxib Ibuprofen
Supplementary analgesic <2 h (n) 72% (13) 68% (27) 43% (17) 0.03
Time to supplementary analgesia (min) 62 (35250) 62 (35278) 156 (55300) 0.09
Pain score at 4 h 20 (040) 20 (1040) 20 (1840) 0.89
Vomited (n) 22% (4) 33% (13) 28% (11) 0.71
Antiemetic administered (n) 22% (4) 20% (8) 20% (8) 0.97
Primary haemorrhage (n) 5.5% (1) 2.5% (1) 5% (2) 0.83
Secondary haemorrhage (n) 11% (2) 7.5% (3) 2.5% (1) 0.30
Pickering et al.
74
in the early postoperative period in the ibuprofen group.
This analgesic benet from the combination of paracetamol
and ibuprofen lends support to our current practice of using
NSAIDs for perioperative analgesia.
Paracetamol 20 mg kg
1
alone provided adequate anal-
gesia for just 24% of the children in this study. This
emphasizes the difculty in providing pain relief in children
with contraindications to NSAIDs. A previous study of
paediatric tonsillectomy has shown paracetamol
(20 mg kg
1
) to be equivalent to diclofenac (1 mg kg
1
),
but each drug alone was noted to provide inadequate overall
analgesia.
11
These results provide some justication for the
use of NSAIDs in children with relative contraindications,
such as controlled asthmatics, particularly in view of the
recent demonstration of the safety of diclofenac in asthmatic
children.
22
However, it should be noted that, although the
dose of paracetamol (20 mg kg
1
) used in this study was our
usual oral loading dose, recent studies have demonstrated
improved analgesic efcacy with higher doses of oral
(40 mg kg
1
) or rectal paracetamol (4060 mg kg
1
), which
achieve higher plasma concentrations.
9 10
It is possible that
the use of a larger loading dose of paracetamol in our study
would have produced an improved analgesic effect. Also
of relevance to our study is the observation that the
premedication of children having adenotonsillectomies with
oral paracetamol elixir (40 mg kg
1
) has been shown to have
no effect on the volume or pH of gastric contents at the time
of operation.
23
The addition of rofecoxib to paracetamol failed to confer
any additional analgesic benet compared with paracetamol
alone. This was a surprising nding, which questions the
role of COX-2 antagonists for perioperative analgesia.
Previous perioperative studies in adults have demonstrated
rofecoxib to have an analgesic efcacy equivalent to that of
ibuprofen
17 18
and it has been shown to be opioid-sparing
after spinal surgery.
24
When designing our study, in the
absence of paediatric pharmacokinetic data, we calculated
the dose of rofecoxib (0.625 mg kg
1
) such that a 40-kg
child would receive the current UK maximum adult dose of
25 mg. Our rofecoxib dose calculation was based on the
nding that, in adults, a dose of rofecoxib 50 mg is
equianalgesic to 400 mg ibuprofen.
17
As the usual paediatric
dose of ibuprofen is 5 mg kg
1
, we expected a dose of
rofecoxib 0.625 mg kg
1
to be equianalgesic. This dose also
allowed us to give equal volumes of the ibuprofen and
rofecoxib drug syrups (0.25 ml kg
1
), which assisted with
study blinding.
The study drugs were given more than 60 min before
surgery, allowing approximately 100 min for rofecoxib
absorption and the onset of analgesic action before arrival in
the recovery unit. Data from dental extraction studies have
indicated that the time to onset of analgesia for rofecoxib is
similar to that of ibuprofen
18
and has been quoted by the
manufacturer as 45 min. We have no evidence of a delayed
analgesic effect of rofecoxib as the total analgesic con-
sumption in the rst 24 h was similar across the groups. In
seeking to explain our failure to demonstrate any benet, it
is unlikely that our study was under-powered, as we have
been able to show the benecial analgesic effect of
ibuprofen.
This study looked for an additive analgesic effect of the
combination of rofecoxib and paracetamol. The sites of
action of both drugs may be relevant: paracetamol inhibits
brain cyclo-oxygenase
25
and rofecoxib inhibits COX-2,
which is expressed constitutively in the brain as well as
being inducible in the periphery.
16
If both drugs were acting
to inhibit the same brain cyclo-oxygenase, we could perhaps
explain the lack of additive effect (if we assume that
paracetamol is producing nearly maximal antagonism of
brain COX). Alternatively, it is possible that inducible
Fig 1 Surgical blood loss. Box plots (showing 10th, 25th, 50th, 75th and 90th centiles) of intraoperative blood loss for children having tonsillectomies
and adenotonsillectomies according to treatment group. There were no signicant differences between the groups.
Analgesia for paediatric tonsillectomy
75
COX-2 does not play a major role in nociception in the early
recovery period after tonsillectomy. Our ndings suggest
that central cyclo-oxygenase and peripheral COX-1 are
involved in mediating early pain after tonsillectomy.
The surgical blood loss seen during this study was similar
to that reported previously from our institution.
13
There
were no signicant differences in blood loss or in bleeding
complications between the three groups. In designing the
study, we wished to reect our normal clinical practice so no
restriction was placed upon surgical grade or technique
employed for tonsillectomy. A retrospective power analysis
showed that the study had limited power (23%) to demon-
strate the largest difference seen between the groups
because of the large variation in blood loss. That the
observed distributions were so wide suggests that there are
factors other than the perioperative use of NSAIDs that
determine surgical bleeding.
The incidence of postoperative vomiting was 29% in the
rst 24 h. This was without prophylactic antiemetic and is
lower than that previously reported in this surgical popu-
lation.
26
Previous studies of tonsillectomy have correlated
postoperative vomiting with opioid use
3
and pain.
11
In our
study, there were no signicant differences between the
groups in vomiting or in the need for rescue antiemetic.
In conclusion, this clinical study emphasizes that
paediatric tonsillectomy is a painful procedure requiring
proactive analgesic strategies to provide a good quality of
postoperative recovery. We have shown analgesic benet
without increased risk of complications by combining
ibuprofen with paracetamol for premedication. We failed
to show an analgesic effect of rofecoxib in combination with
paracetamol, suggesting that the COX-2 antagonists will
require careful evaluation before their role in perioperative
analgesia can be dened clearly. In particular, studies that
compare COX-2 antagonists with paracetamol (which,
coincidentally, also has a low incidence of gastric and
haemostatic complications) will be particularly valuable.
There is further scope for improving analgesia after
tonsillectomy but we can commend the perioperative
combination of ibuprofen with paracetamol as a useful
strategy in children.
Acknowledgements
We would like to thank the nursing staff and anaesthetists at Bristol Royal
Hospital for Children for their help with the conduct of this study. We are
most grateful to Rosemary Greenwood, RDSU, UBHT for statistical
analysis and to the pharmacy department at UBHT for assistance with
randomization and preparation of study drugs.
References
1 Morton NS. Prevention and control of pain in children. Br J
Anaesth 1999; 83: 11829
2 Semple D, Russell S, Doyle E, Aldridge LM. Comparison of
morphine sulphate and codeine phosphate in children
undergoing adenotonsillectomy. Paediatr Anaesth 1999; 9: 1358
3 Mather SJ, Peutrell JM. Postoperative morphine requirements,
nausea and vomiting following anaesthesia for tonsillectomy.
Comparison of intravenous morphine and nonopioid analgesic
techniques. Paediatr Anaesth 1995; 5: 1858
4 Watters CH, Patterson CC, Mathews HML, Campbell W.
Diclofenac sodium for post-tonsillectomy pain in children.
Anaesthesia 1988; 43: 6413
5 Bone ME, Fell D. A comparison of rectal diclofenac with
intramuscular papaveretum or placebo for pain relief following
tonsillectomy. Anaesthesia 1988; 43: 27780
6 Splinter WM, Rhine EJ, Roberts DW, Reid CW, MacNeill HB.
Preoperative ketorolac increases bleeding after tonsillectomy in
children. Can J Anaesth 1996; 43: 5603
7 Rusy LM, Houck CS, Sullivan LJ, et al. A double-blind evaluation
of ketorolac tromethamine versus acetaminophen in pediatric
tonsillectomy: analgesia and bleeding. Anesth Analg 1995; 80:
2269
8 Romsing J, Walther-Larsen S. Peri-operative use of nonsteroidal
anti-inammatory drugs in children: analgesic efcacy and
bleeding. Anaesthesia 1997; 52: 67383
9 Anderson B, Kanagasundarum S, Woolard G. Analgesic efcacy
of paracetamol in children using tonsillectomy as a pain model.
Anaesth Intens Care 1996; 24: 66973
10 Korpela MD, Korvenoja P, Meretoja OA. Morphine-sparing
effect of acetaminophen in pediatric day-case surgery.
Anesthesiology 1999; 91: 4427
11 Romsing J, Ostergaard D, Drozdziewicz D, Schultz P, Ravn G.
Diclofenac or acetaminophen for analgesia in paediatric
tonsillectomy outpatients. Acta Anaesthiol Scand 2000; 44: 2915
12 Kehlet H, Dahl JB. The value of `multimodal' or `balanced
analgesia' in postoperative pain treatment. Anesth Analg 1993; 77:
104856
13 Mendham JE, Mather SJ. Comparison of diclofenac and
tenoxicam for postoperative analgesia with and without
fentanyl in children undergoing adenotonsillectomy or
tonsillectomy. Paediatr Anaesth 1996; 6: 46773
14 Morton NS, O'Brien K. Analgesic efcacy of paracetamol and
diclofenac in children receiving PCA morphine. Br J Anaesth 1999;
82: 7157
15 Montgomery JE, Sutherland CJ, Kestin IG, Sneyd JR. Morphine
consumption in patients receiving rectal paracetamol and
diclofenac alone and in combination. Br J Anaesth 1996; 77: 4457
16 Hawkey CJ. COX-2 inhibitors. Lancet 1999; 353: 30714
17 Morrison BW, Christensen S, Yuan W, Brown J, Amlani S,
Seidenberg B. Analgesic efcacy of the cyclooxygenase-2-specic
inhibitor rofecoxib in post-dental surgery pain: a randomised,
controlled trial. Clin Ther 1999; 21: 94353
18 Ehrich EW, Dallob A, De Lepeleire I, et al. Characterization of
rofecoxib as a cyclooxygenase-2 isoform inhibitor and
demonstration of analgesia in the dental pain model. Clin
Pharmacol Ther 1999; 65: 33647
19 Wolf AR. Tears at bedtime: a pitfall of extending paediatric day-
case surgery without extending analgesia. Br J Anaesth 1999; 82:
31920
20 Beyer JE, Denyes MJ, Villarruel AM. The creation, validation, and
continuing development of the Oucher: a measure of pain
intensity in children. J Pediatr Nurs 1992; 7: 33546
21 Sutters KA, Levine JD, Dibble S, Savedra M, Miaskowski C.
Analgesic efcacy and safety of single-dose intramuscular
ketorolac for postoperative pain management in children
following tonsillectomy. Pain 1995; 61: 14553
22 Short JA, Barr CA, Palmer CD, Goddard JM, Stack CG, Primhak
RA. Use of diclofenac in children with asthma. Anaesthesia 2000;
55: 3347
23 Anderson BJ, Rees SG, Liley A, Stewart AW, Wardill MJ. Effect of
Pickering et al.
76
preoperative paracetamol on gastric volumes and pH in children.
Paediatr Anaesth 1999; 9: 2037
24 Reuben SS, Connelly NR. Postoperative analgesic effects of
celecoxib or rofecoxib after spinal fusion surgery. Anesth Analg
2000; 91: 12215
25 Flower RJ, Vane JR. Inhibition of prostaglandin synthetase in brain
explains the antipyretic activity of paracetamol (4-acetamido-
phenol). Nature 1972; 240: 4101
26 Jensen AB, Christiansen DB, Coulthard K, et al. Tropisetron
reduces postoperative vomiting in children undergoing
tonsillectomy. Paediatr Anaesth 2000; 10: 6975
Analgesia for paediatric tonsillectomy
77