Journal of Clinical Anesthesia (2015) xx, xxxxxx

Original contribution

Tramadol wound infiltration is not different from

intravenous tramadol in children: a randomized
controlled trial
Ana Laura Albertoni Giraldes MD (Anesthetist)a ,
Angela Maria Sousa MD, PhD (Chief)b,, Alexandre Slullitel MD, MSc (Anesthetist)c ,
Gabriel Magalhes Nunes Guimares MD (Anesthetist) b ,
Melina Genevive Mary Egan Santos MD (Anesthetist)d ,
Renata Evangelista Pinto MD (Anesthetist)d , Hazem Adel Ashmawi MD, PhD (Anesthetist)e ,
Rioko Kimiko Sakata MD, PhD (Assistant Professor)a
a

Department of Anesthesia, Surgery Division, Pediatric Anesthesia Unit, Federal University of So Paulo, So Paulo, Brazil
Pain Management Service, Cancer Institute of the State of Sao Paulo, So Paulo, Brazil
c
Department of Anesthesia and Pain Management, Santa Paula Hospital, So Paulo, Brazil
d
Department of Anesthesia, University Hospital, University of So Paulo, So Paulo, Brazil
e
Department of Anesthesia, Hospital das Clinicas from the University of So Paulo, So Paulo, Brazil
b

Received 5 December 2014; revised 8 May 2015; accepted 13 August 2015

Keywords:
Infiltration anesthesia;
Tramadol;
Intravenous administration;
Postoperative pain;
Hernioplasty

Abstract
Study Objective: The purpose of this trial was to assess if tramadol wound infiltration is superior to
intravenous (IV) tramadol after minor surgical procedures in children because tramadol seems to have
local anestheticlike effect.
Design: Randomized double-blind controlled trial.
Setting: Postanesthesia care unit.
Patients: Forty children, American Society of Anesthesiologists physical status I or II, scheduled to
elective inguinal hernia repair.
Interventions: Children were randomly distributed in 1 of 2 groups: IV tramadol (group 1) or
subcutaneous infiltration with tramadol (group 2). At the end of the surgery, group 1 received 2 mg/kg
tramadol (3 mL) by IV route and 3-mL saline into the surgical wound; group 2 received 2 mg/kg
tramadol (3 mL) into the surgical wound and 3-mL saline by IV route.
Measurements: In the postanesthesia care unit, patients were evaluated for pain intensity, nausea and
vomiting, time to first rescue medication, and total rescue morphine and dipyrone consumption.
Main Results: Pain scores measured during the postanesthesia recovery time were similar between
groups. Time to first rescue medication was shorter, but not statistically significant in the IV group.
The total dose of rescue morphine and dipyrone was also similar between groups.

Disclosure: No fund sources.

Corresponding author at: Angela Maria Sousa, Department of Anesthesia, Hospital das Clinicas from the University of Sao Paulo. Av. Dr Enas de
Carvalho Aguiar, 255 - 8th floor - PAMB - Cerqueira Csar - CEP 05403-000, Brazil. Tel.: + 551126616680; fax: + 551138850997.
E-mail address: angela.sousa0309@gmail.com (A.M. Sousa).
http://dx.doi.org/10.1016/j.jclinane.2015.08.009
0952-8180/ 2015 Elsevier Inc. All rights reserved.

A.L.A. Giraldes et al.

Conclusions: We concluded that tramadol was effective in reducing postoperative pain in children, and
there was no difference in pain intensity, nausea and vomiting, or somnolence regarding IV route or
wound infiltration.
2015 Elsevier Inc. All rights reserved.

1. Introduction
It is well recognized that optimal postoperative pain
management is essential for children and that analgesia
should be started even before the surgery. Standardized
approaches for postoperative nausea and vomiting and pain
control are both important factors to be optimized in
outpatient children undergoing minor surgical procedures.
Preventive multimodal analgesia, which comprises a combination of different drugs and regional anesthesia techniques, can offer the advantage of high-quality pain relief [1]
and reduced amount of opioid consumption in the perioperative period [2]. Local anesthetics administered either by
neuraxial techniques, peripheral nerve blocks, or local
infiltration are adopted as adjuvants to general anesthesia
and are associated to excellent pain control, rapid recovery
from anesthesia, and fewer opioid side effects [3].
Tramadol has been demonstrated to have local analgesic
effect in previous studies in adults [4,5] and children [6-9].
Besides being a weak synthetic opioid, tramadol inhibits the
reuptake of monoaminergic neurotransmitters (5-hydroxytryptamine and noradrenaline) and has a local anesthetic
like action on peripheral nerves [10] similar to lidocaine 1%
[11]. The addition of tramadol to local anesthetics in peripheral
nerve block prolongs the duration of the analgesia [12,13]
and has an analgesic effect similar to levobupivacaine when
injected subcutaneously [4].
The objective of this study was to evaluate if wound
infiltration with tramadol confers additional beneficial
results in terms of postoperative pain control of children
submitted to minor pediatric surgeries.

2. Subjects and methods

After approval by institutional ethics committees for
Analysis of Research Projects of Darcy Vargas Children's
Hospital and So Paulo Hospital of the Federal University
of So Paulo and written informed parental consent had
been obtained, children were scheduled to undergo unilateral
inguinal hernia repair.
This prospective double-blind randomized clinical trial
consecutively recruited 40 American Society of Anesthesiologists (ASA) physical status I or II children, aged between 3
and 12 years, undergoing elective unilateral inguinal hernia
repair. Exclusion criteria were parental refusal, preoperative
pain, previous analgesic ingestion in the 48-hour period

before surgery, known or suspected allergic reaction to any

of the drugs studied, or previous history of coagulopathy.
Children were then randomized (random number generator) preoperatively using sealed envelopes to receive 1 of
the 2 treatments: a dose of 2 mg/kg of tramadol in a total
3-mL volume syringe either by intravenous (IV) route (group
1) or by wound infiltration (group 2). The initial tramadol
concentration was 50 mg/mL, and it was further diluted to a
maximum 3-mL saline solution, so final concentration varied
according to patient's weight. An independent nurse not
involved in the study prepared both administered solutions.
At the time of skin suture, according to the allocation group,
1 of these 2 different solutions was blindly made available
for the anesthesiologist to be administered by IV route,
whereas the other was equally provided to the surgical team
to perform wound infiltration.
Children were given 0.5 mg/kg oral midazolam 30
minutes before anesthesia induction, which consisted in
4% to 5% end-tidal sevoflurane concentration in 100%
oxygen. Standard monitoring consisted of electrocardiogram, pulse oximetry, capnography, and noninvasive
continuous blood pressure measurement. Once an adequate
depth of anesthesia had been achieved, an IV access was
secured. A laryngeal mask airway was used to manage the
airways. Before incision, ilioinguinal-iliohypogastric nerve
blockade (IINB) was performed according to classical
anatomical landmarks. The anterior superior iliac spine was
palpated, and a point 1.0 to 1.5 cm cephalad and toward the
midline was located. A 22-gauge short beveled needle was
passed through the external and internal oblique muscles,
and 2.0 to 2.5 mL of local anesthetic was deposited in a
fan-like fashion cephalad toward the umbilicus, medially,
and caudad toward the groin. As the needle was advanced
through the external and internal oblique muscles, a pop was
elicited providing a guide of proper needle placement. Just
before removal from the skin, another 0.5 to 1.0 mL of local
anesthetic was injected subcutaneously to block the
iliohypogastric nerve. So, a total of 3 mL of 1% lidocaine
solution was injected into these 2 layers. Anesthesia was
maintained with 50% oxygen/air mixture and sevoflurane
(2.5%-3%) with spontaneous breathing throughout the
procedure. After incision, IV fentanyl (0.5 g/kg) was
administered if blood pressure or heart rate had more than
30% increase compared to preoperative values. If necessary,
assisted ventilation was provided according to end-tidal
carbon dioxide measurements. If a second fentanyl rescue
dose was considered necessary, the ilioinguinal nerve block
has judged insufficient, and the patient was excluded from

Tramadol wound infiltration in children

the study. Basal values of heart rate (HR), systolic blood
pressure, diastolic blood pressure, mean arterial pressure,
end-tidal carbon dioxide, and peripheral oxygen saturation
were recorded just before induction and at each 5-minute
interval till the end of the surgery.
At the end of the procedure, anesthesia was discontinued,
Laringeal Mask Airway was removed, and the patient was
left to recover in the postanesthesia care unit (PACU).
Pain was evaluated using 2 different scales in the PACU.
Six-point Wong Baker Faces Pain Rating Scale (WBFPRS)
(0 = no hurt; 1 = hurts little bit; 2 = hurts little more; 3 = hurts
even more; 4 = hurts whole lot; 5 = hurts worst) and Simple
Descriptive Scale (SDS) (0 = no pain; 1 = mild pain; 2 =
moderate pain; 3 = severe pain). Pain at rest was evaluated
soon after awake from anesthesia, then at 10, 20, 60, and 120
minutes after the end of the surgery.
When pain was scored greater than 4 by the WBFPRS,
moderate by SDS, or continuous crying persisted longer than
5 minutes, a first IV dipyrone bolus dose (30 mg/kg) was
administered by an anesthesiologist blinded to the study
intervention. If pain persisted for more than 10 minutes after
first IV dipyrone administration, morphine (0.1 mg/kg) was
provided, and the time to first rescue medication was noted.
After establishment of oral feeding, patients were
discharged from the hospital, and parents were oriented to
contact the researcher in case of uncontrolled pain, nausea, or
vomiting. Medical appointment was scheduled 1 week later.

2.1. Outcomes
The primary outcome was pain intensity assessed
according to pain scores, time to first analgesic rescue
administration, total dose of rescue dipyrone (in milligrams
per kilogram), and total dose of rescue morphine (in
milligrams per kilogram) for the first 120 minutes after the
end of the procedure. Side effects such as nausea, vomiting,
and sedation were considered as secondary outcomes.
Sedation was assessed by a Ramsay Scale and was
considered at least mild when patients scored 4 or higher (1 =
patient awake and anxious, agitated, or restless; 2 = patient
awake and cooperative, oriented and calm; 3 = patient asleep,
responsive to commands.; 4 = patient asleep, with brisk
response to stimuli [light and noise]; 5 = patient asleep, with
response only to pain; 6 = patient with no response to any
stimuli [light, noise, or pain]).
Postoperative pain and common side effects were
assessed by an anesthesiologist trained in pediatric anesthesia
and pain management.
Nausea and vomiting were registered according to their
occurrence during PACU stay before discharge to the ward.

2.2. Statistical analysis

Considering that a 1-point difference in the mean of
6-point WBFPRS score between the 2 groups is clinically

3
relevant and that the SD of WBFPRS based on a pilot study
in our institution was 0.8, to achieve a power of 95% with a type I
error rate of 0.05 for the Student t test, 17 patients per study group
were necessary. So, we included 20 patients in each group.
To assess the null hypothesis, Fisher exact test or
Wilcoxon-Mann-Whitney for ordinal data tests using R
software (R Foundation for statistical computing, Vienna,
Austria) were used.

3. Results
A total of 40 patients were enrolled in the study. Two
patients in group I were excluded due to venous line
extravasation, 1 was excluded for inadequate nerve block
and 1 patient in group II was excluded due to intravenous
line extravasation. Overall data from 37 patients19 in
group I (IV tramadol) and 18 in group II (tramadol wound
infiltration)were analyzed.
There was no difference in age, weight, surgery duration,
ASA physical status, and total amount of fentanyl consumption during anesthesia between the 2 groups (Table 1). Both
groups were similar comparing MAP and HR, both in
operating room and in PACU (data not shown).
Pain scores measured by WBFPRS and SDS were similar
between groups. Time to first rescue medication was shorter
in group 1 (IV group), but it did not reach statistical
significance (Table 2; P = .77). The total dose of rescue
morphine and dipyrone was also similar between groups (P =
.35) (Table 2).
There were no statistically differences in side effects such
as nausea, vomiting, and sedation between groups (Fisher
exact test; Table 3).

4. Discussion
Pain is a highly individualized and subjective event,
influenced by cultural and emotional aspects. Because
assessment of pain is difficult in clinical practice, anesthesia
planning should begin with preoperative assessment of
anxiety of parents and children. A multimodal and
preventive analgesia approach, in which nonopioid analgesics, such as nonsteroidal anti-inflammatory drugs and local
anesthetics are combined with opioids, is recommended to
maximize pain control and minimize drug-induced adverse
side effects [14].
The optimal analgesic strategy for pediatric inguinal
hernia repair remains undefined. Available evidence comparing caudal blockade to alternative analgesic strategies in
achieving postoperative analgesia is still controversial [15].
However, the addition of an ultrasound-guided ilioinguinal
nerve block to a single-shot caudal block is efficient to
decrease pain scores in inguinal hernia repair patients [16].

T1

T2

T3

A.L.A. Giraldes et al.

In our institution, peripheral nerve blocks are associated

to general anesthesia in most of the patients because the
availability of potent analgesic medications labeled for
children use has been limited. Our option for using tramadol
in this study is based on the fact it has been proven safe, with
irrelevant side effects in children [17], although it still needs
good scientific evidence [18]. Previous experimental
[10,11,19] and clinical studies [6,20], comparing tramadol
infiltration with local anesthetics, showed that this procedure
resulted in local anesthesia for minor surgeries in adults [6,7]
and children [7,20].
Based on these observations, we planned to administer
tramadol via wound to acquire local effect of the parent
drug. Therefore, we got similar analgesia whether intravenous or tramadol wound infiltration was used in children
undergoing hernia repair. Any remarkable difference was
also found in the incidence of side effects, such as nausea,
vomiting, and sedation.
These results may be explained by a combination of
factors, from pharmacological effects of the drugs used
(lidocaine or tramadol) to a wide age bracket (3-12 years old)
variation of the population studied.
The first aspect to be observed is tramadol pharmacokinetic characteristics. Tramadol needs first pass metabolism to
act as opioid analgesic, whereas the local anesthetic effect
happens immediately after subcutaneous injection [10,19].
Although the local anesthetic effect induced by the parent
drug initiates analgesia, it gives time to tramadol to be
absorbed and its metabolites exert the systemic effect. The
local effect associated to the systemic effect (Tmax after
subcutaneous injection 20.6 minutes, with a half-life of 5.2
hours) [21] could bring additional benefits of tramadol use in
postoperative pain analgesia. Tramadol pharmacokinetics
after intravenous administration is similar in adults and
children [22]. However, tramadol pharmacokinetics after
subcutaneous route administration has never been reported
in children.
Pain assessment in children is one of the most difficult
challenges that researchers and health professionals face.
Based on previous reports, [5,20,23], early analgesic effect
of tramadol wound infiltration was expected in the first
Table 1
Variable

Patients characteristics and surgery duration.

Group I,
mean (SD)
or prevalence

Group II,
mean (SD)
or prevalence

Female/male (%)
0.59
0.61
Age (mo)
57.15 (30.28) 70.66 (28.72)
Weight (kg)
20.18 (5.51) 23.81 (9.32)
ASA
I
0.85
0.95
II
0.15
0.05
Surgery duration (min) 34.21 (14.10) 33.33 (15.22)
Total fentanyl (g/kg)
0.12 (0.40)
0.12 (0.25)

1
.3863
.2675
.60

.891
.4631

Group I, intravenous tramadol; group II, tramadol wound infiltration.

Table 2

Main outcomes.

Outcome

Group I Group II P*
(median) (median)

WBFPRS
5 min
10 min
20 min
60 min
120 min
Simple descriptive pain scale
5 min
10 min
20 min
60 min
120 min
Total dose (mg/kg) of rescue drug,
mean (DP)
Morphine
Dipyrone
Time to first analgesic demand
(min), mean (DP)

1.5
3
0
0
0

2
1
1
0
0

.9037
.2278
.4582
.609
.3126

2
2
0
0
0

1
1.5
0
0
0

.6971
.8379
.4315
.132
.2899

0.041
(0.04)
22
(17.52)
63.0
(17.20)

0.054
(0.06)
21.5
(15.72)
70.6
(32.8)

.683
.9141
.7705

Mann-Whitney U test. Pb0.05 was considered significant.

Group I, intravenous tramadol; group II, wound infiltration with tramadol.

minutes after the end of the surgery, just after admission in

the PACU, when children were evaluated by faces scale. It
should be noticed that, in the first 10 to 15 minutes after the
anesthesia awakening, younger children tended to be more
tearful, and they may have been on the extremes of an all or
nothing context. In fact, young children do not always
understand that the scale may represent a continuum score
system, going from no pain to worst possible pain [24], and
caregivers must be led to wrong conclusions concerning the
analgesic effect of tramadol, based only on faces scale [25].
This is the most popular approach to elicit children
self-report of pain [26], although not a perfect instrument.
Besides what has been said, the tramadol dose (2 mg/kg)
injected by IV or wound infiltration we used was similar to
the one in previous studies [5,20,23] that showed only local
effect of the drug. In our study, we would like to show both
local and systemic effects. The difference of age in the
population studied (3-12 years old) generated different drug
concentration because it depends on the weight of the child.

Table 3

Secondary outcomes.

Outcome

Group I
(prevalence)

Group II
(prevalence)

Nausea
Vomiting
Somnolence

0.05
0.05
0.42

0.11
0.11
0.33

.6039
1
.7374

Group I, intravenous tramadol; group II, wound infiltration with

tramadol.

Tramadol wound infiltration in children

This factor has to be taken into consideration when we
investigate the local effect of tramadol. In addition,
subcutaneous absorption is conditioned to the ionization
constant of tramadol solution, which may be affected by the
dilution of the drug in different concentrations, which may
interfere with the parent drug systemic absorption.
The last consideration to be made is the ilioinguinal nerve
blockade before surgical incision with 1% lidocaine. Blood
concentrations of local anesthetics in young children after
ilioinguinal block can be twice those measured in older
children [16], which would make local anesthesia duration
extremely variable in the studied population [27]. However,
it has been reported that IINB could provide analgesia for 60
to 120 minutes [19,20]. As a result, local analgesic effect of
tramadol could not be exclusively demonstrated in the
present research. On the other hand, for both groups
(intravenous or subcutaneous), predicted longer pharmacokinetic profile of tramadol may have overcome the effect of
the IINB provided by lidocaine itself.
In our opinion, the major obstacles of this study are the
limitation of young children to self-report pain, the lack of a
placebo group, and the peripheral nerve blockade before
surgical incision. We chose not to change the anesthesia
protocol but prioritize children comfort. Further studies
concerning local effect of drugs should use patient controlled
analgesia as an analgesic titration method and separate
children younger than 4 years old from older ones, to get a
more effective feedback of pain evaluation.

[8]

[9]

[10]

[11]

[12]

[13]

[14]
[15]

[16]

[17]

Acknowledgments

[18]

The authors thank Darcy Vargas Children's Hospital for

providing the conditions for this research.

[19]

[20]

References
[1] Lnnqvist P-A. Blocks for pain management in children undergoing
ambulatory surgery. Curr Opin Anaesthesiol 2011;24:627-32.
[2] Ecoffey C. Safety in pediatric regional anesthesia. Paediatr Anaesth
2012;22:25-30.
[3] Boretsky KR. Regional anaesthesia in paediatrics: marching forward.
Curr Opin Anaesthesiol 2014;27:556-60.
[4] Demiraran Y, Albayrak M, Yorulmaz IS, Ozdemir I. Tramadol and
levobupivacaine wound infiltration at cesarean delivery for postoperative
analgesia. J Anesth 2013;27:175-9.
[5] Ozyilmaz K, Ayoglu H, Okyay RD, Yurtlu S, Koksal B, Hanci V, et al.
Postoperative analgesic effects of wound infiltration with tramadol and
levobupivacaine in lumbar disk surgeries. J Neurosurg Anesthesiol
2012;24:331-5.
[6] Karg E, Ikdemir A, Tokgz H, Erol B, Iikdemir F, Hanci V, et al.
Comparison of local anesthetic effects of tramadol with prilocaine
during circumcision procedure. Urology 2010;75:672-5.
[7] Heiba MH, Atef A, Mosleh M, Mohamed R, El-Hamamsy M.
Comparison of peritonsillar infiltration of tramadol and lidocaine for

[21]

[22]
[23]

[24]

[25]

[26]

[27]

the relief of post-tonsillectomy pain. J Laryngol Otol 2012;126:

1138-41.
Polat F, Tuncel A, Balci M, Aslan Y, Sacan O, Kisa C, et al.
Comparison of local anesthetic effects of lidocaine versus tramadol
and effect of child anxiety on pain level in circumcision procedure. J
Pediatr Urol 2012;9:670-4.
Sezen G, Demiraran Y, Karagoz I, Kucuk A. The assessment of
bupivacaine-tramadol and levobupivacaine-tramadol combinations for
preemptive caudal anaesthesia in children: a randomized, double-blind,
prospective study. Int J Clin Exp Med 2014;7:1391-6.
Katsuki R, Fujita T, Koga A, Liu T, Nakatsuka T, Nakashima M, et al.
Tramadol, but not its major metabolite (monoOdemethyl tramadol)
depresses compound action potentials in frog sciatic nerves. Br J
Pharmacol 2006;149:319-27.
Sousa A, Ashmawi H, Costa L, Posso IP, Slullitel A. Percutaneous
sciatic nerve block with tramadol induces analgesia and motor
blockade in two animal pain models. Braz J Med Biol Res 2012;45:
147-52.
Kaabachi O, Ouezini R, Koubaa W, Ghrab B, Zargouni A, Ben
Abdelaziz A. Tramadol as an adjuvant to lidocaine for axillary brachial
plexus block. Anesth Analg 2009;108:367-70.
Allemanno F, Ghisi D, Fanelli A, Faliva A, Pergolotti B, Bizzarri F,
et al. Tramadol and 0.5% levobupivacaine for single-shot interscalene
block: effects on postoperative analgesia in patients undergoing
shoulder arthroplasty. Minerva Anestesiol 2012;78:291-6.
Yaster M. Multimodal analgesia in children. Eur J Anaesthesiol 2010:
851-7.
Baird R, Guilbault MP, Tessier R, Ansermino JM. A systematic review
and meta-analysis of caudal blockade versus alternative analgesic
strategies for pediatric inguinal hernia repair. J Pediatr Surg 2013;48:
1077-85.
Jagannathan N, Sohn L, Sawardekar A, Ambrosy A, Hagerty J, Chin A,
et al. Unilateral groin surgery in children: will the addition of an ultrasoundguided ilioinguinal nerve block enhance the duration of analgesia of a
single-shot caudal block? Paediatr Anaesth 2009;19:892-8.
Bozkurt P. Use of tramadol in children. Paediatr Anaesth 2005;15:
1041-7.
Schnabel A, Reichl SU, Meyer-Frieem C, Zahn PK, Pogatzki-Zahn E.
Tramadol for postoperative pain treatment in children. Cochrane
Database Syst Rev 2015:CD009574.
Sousa AM, Ashmawi HA. Local analgesic effect of tramadol is not
mediated by opioid receptors in early postoperative pain in rats. Braz J
Anesthesiol 2015;65:186-90.
Kargi E, Babuccu O, Altunkaya H, Hosnuter M, Ozer Y, Babuccu B,
et al. Tramadol as a local anaesthetic in tendon repair surgery of the
hand. J Int Med Res 2008;36:971-8.
Murthy BV, Pandya KS, Booker PD, Murray A, Lintz W, Terlinden R.
Pharmacokinetics of tramadol in children after i.v. or caudal epidural
administration. Br J Anaesth 2000;84:346-9.
Saudan S, Habre W. Pharmacokinetics of tramadol in children. Ann Fr
Anesth Reanim 2007;26:560-3.
Matkap E, Bedirli N, Akkaya T, Gm H. Preincisional local infiltration
of tramadol at the trocar site versus intravenous tramadol for pain control
after laparoscopic cholecystectomy. J Clin Anesth 2011;23:197-201.
Hicks CL, von Baeyer CL, Spafford PA, Korlaar I, Goodenough B.
The faces scale revised: toward a common metric pediatric pain
measurement. Pain 2001;93:173-83.
Vervoort T, Caes L, Crombez G, Koster E, Van Damme S, Dewitte M,
et al. Parental catastrophizing about children's pain and selective
attention to varying levels of facial expression of pain in children: a
dot-probe study. Pain 2001;152:1751-7.
Tomlinson D, von Baeyer CL, Stinson JN, Sung L. A systematic
review of faces scales for the self-report of pain intensity in children.
Pediatrics 2010;126:e1168-98.
Abdellatif AA. Ultrasound-guided ilioinguinal/iliohypogastric nerve
blocks versus caudal block for postoperative analgesia in children
undergoing unilateral groin surgery. Saudi J Anaesth 2012;6:367-72.