Moore et al.

The Journal of Headache and Pain (2015) 16:60

DOI 10.1186/s10194-015-0541-5

RESEARCH ARTICLE Open Access

Dexketoprofen/tramadol: randomised
double-blind trial and confirmation of empirical
theory of combination analgesics in acute pain
R. Andrew Moore1*, C. Gay-Escoda2, R. Figueiredo2, Z. Tóth-Bagi3, T. Dietrich4, S. Milleri5, D. Torres-Lagares6,
C. M. Hill7, A. García-García8, P. Coulthard9, A. Wojtowicz10, D. Matenko10, M. Peñarrocha-Diago11, S. Cuadripani12,
B. Pizà-Vallespir12, C. Guerrero-Bayón12, M. Bertolotti13, M. P. Contini13, S. Scartoni13, A. Nizzardo13, A. Capriati13
and C. A. Maggi13

Abstract
Background: Combination analgesics are effective in acute pain, and a theoretical framework predicts efficacy for
combinations. The combination of dexketoprofen and tramadol is untested, but predicted to be highly effective.
Methods: This was a randomised, double-blind, double-dummy, parallel-group, placebo-controlled, single-dose trial
in patients with moderate or severe pain following third molar extraction. There were ten treatment arms, including
dexketoprofen trometamol (12.5 mg and 25 mg) and tramadol hydrochloride (37.5 mg and 75 mg), given as four
different fixed combinations and single components, with ibuprofen 400 mg as active control as well as a placebo
control. The study objective was to evaluate the superior analgesic efficacy and safety of each combination and
each single agent versus placebo. The primary outcome was the proportion of patients with at least 50 % max
TOTPAR over six hours.
Results: 606 patients were randomised and provided at least one post-dose assessment. All combinations were
significantly better than placebo. The highest percentage of responders (72 %) was achieved in the dexketoprofen
trometamol 25 mg plus tramadol hydrochloride 75 mg group (NNT 1.6, 95 % confidence interval 1.3 to 2.1).
Addition of tramadol to dexketoprofen resulted in greater peak pain relief and greater pain relief over the longer
term, particularly at times longer than six hours (median duration of 8.1 h). Adverse events were unremarkable.
Conclusions: Dexketoprofen trometamol 25 mg combined with tramadol hydrochloride 75 mg provided good
analgesia with rapid onset and long duration in a model of moderate to severe pain. The results of the dose
finding study are consistent with pre-trial calculations based on empirical formulae.
Trial registration: EudraCT (2010-022798-32); Clinicaltrials.gov (NCT01307020).
Keywords: Dexketoprofen; Tramadol; Combination analgesics; Postoperative pain; Third molar; Randomised
controlled trial; Dose range

* Correspondence: [email protected]
1
Pain Research and Nuffield Division of Anaesthetics, Nuffield Department of
Clinical Neurology, University of Oxford, The Churchill, Oxford OX3 7LE, UK
Full list of author information is available at the end of the article

© 2015 Moore et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium,
provided the original work is properly credited.
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 2 of 13

Background Methods
Greater efficacy from combination analgesics in acute The study (Sponsor Code DEX-TRA-02; EudraCT num-
pain has been recognised for some time [1], albeit ber 2010-022798-32) was registered at clinicaltrials.gov
originally in cancer pain. When oral analgesic drugs (NCT01307020). It was performed at 16 study sites in six
are tested in standard, acute pain models [2] those European countries (Germany, Hungary, Italy, Poland,
with the highest efficacy and lowest numbers-needed- Spain and the United Kingdom). It was conducted in
to-treat (NNT) are typically high doses of individual accordance with the principles of Good Clinical Practice
analgesics or low doses of combinations of analgesics and the Declaration of Helsinki and was approved by
[3]. Combinations of drugs of high efficacy include all the concerned Competent Authorities and Ethics
paracetamol and codeine [4], paracetamol and oxycodone Committees. All participating patients provided written
[5], ibuprofen and codeine [6], ibuprofen and oxycodone informed consent. The clinical phase of the study started
[7], and ibuprofen and paracetamol [8]. Even adding on 23rd February 2011 (first patient screened) and
caffeine can improve analgesic efficacy as a combination concluded on 14th October 2011 (last patient out).
with conventional analgesics [9].
The efficacy of combination analgesics has been shown Patients
to be the sum of the efficacies of the individual analgesic Healthy male or female patients, aged 18 to 70 years, were
components, and was broadly true across a range of eligible for the study if they were scheduled for outpatient
different drug combinations, in postoperative pain and surgical removal, under local anaesthesia, of one or more
migraine headache, and when tested in the same and third molars, at least one of which was fully or partially
different trials [10]. This means that the efficacy of any impacted in mandibular bone. Criteria for randomisation
proposed combination can be assessed theoretically before included postoperative pain of moderate to severe intensity
clinical trials are conducted. (Visual Analogue Scale [VAS] ≥40 mm and 4-point Verbal
A potential part of any combination might be a fast- Rating Scale [VRS] ≥2) within four hours after surgery.
acting non-steroidal anti-inflammatory drug (NSAID) Patients were excluded from the study in any of the
formulation, because speed of absorption and onset following circumstances: pregnant or breastfeeding women
produces good and long lasting analgesia [11, 12]. or women of child-bearing potential not using adequate
Dexketoprofen is effective in acute pain at low doses contraception; known allergy to the study drugs, paraceta-
[13], and is also effective in a wide variety of pain con- mol, acetylsalicylic acid, opioids or other NSAIDs; moderate
ditions [14]; dexketoprofen is the active chiral form to severe renal, hepatic or cardiac dysfunction; history of
of ketoprofen. Tramadol is a widely used opioid of gastrointestinal disorders, bleeding disorders; epilepsy,
proven efficacy in combination with paracetamol [15]. asthma, angioedema or related disorders; history of drug or
This study therefore aimed at evaluating the superior alcohol abuse; presence of any medical condition that in the
analgesic efficacy and safety of dexketoprofen trometamol opinion of the investigator might pose a risk to the patient,
and tramadol hydrochloride given as four different fixed may confound study results or might impair compliance
combinations and as single components in comparison to with the study procedures. Patients who had received any
placebo, on moderate to severe acute pain following investigational drug or participated in any other clinical trial
impacted third mandibular molar tooth extraction. It was within the previous month were also excluded. Further
also intended to select the optimum dose combination(s) exclusion criteria included significant surgical complica-
to be further evaluated in the subsequent phase III pivotal tions, overall surgery duration longer than one hour and
studies. need for re-anaesthesia. Patients who had taken any analge-
We used a formula derived empirically to estimate sics less than 24 h before surgery were also excluded.
the possible efficacy that might be obtained from dexketo- Concomitant use of alcohol, psychoactive drugs, sedatives
profen and tramadol dosing combinations [10]. There and any other medications or therapies that could pose a
were limited data for dexketoprofen from a systematic risk to the patient or confound the study results were not
review [13] and two individual patient level analyses permitted within 48 h and two weeks before surgery (de-
of tramadol [16, 17]. Estimated NNTs for combinations pending on the half life of the respective medications) and
with dexketoprofen trometamol 25 mg and tramadol up to 24 h post-dose. Local application of ice and the intake
hydrochloride 37.5 mg or 75 mg were 2.2 and 1.6, of caffeine were not permitted during the 24-hour post-
respectively. Estimates for combinations with lower dose period. Patients had to be in fasting conditions from
dexketoprofen trometamol doses (12.5 mg) were higher two hours before surgery and up to three hours post-dose.
(worse) than 3 or above. There was limited confidence
in the NNT estimates for the combinations due to Study design
uncertainty in the efficacy estimates of individual drugs This was a multicentre, randomised, double-blind, double-
because of low numbers. dummy, parallel-group, placebo-controlled, single-dose,
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 3 of 13

phase II, dose-finding study, with a total of 10 treatment involved in the preparation or the handling of the ran-
arms (with balanced allocation ratio), including dex- domisation list were not involved in the study conduct
ketoprofen trometamol (12.5 mg and 25 mg) and and statistical analysis. Double-blind conditions were
tramadol hydrochloride (37.5 mg and 75 mg) given secured by the identical appearance and weight of the
as four different fixed combinations (DKP12.5/TRAM37.5; eight tested study drugs as tablets as well as the placebo
DKP12.5/TRAM75; DKP25/TRAM37.5; DKP25/TRAM75) tablet matching the tested study drugs. In order to keep
and as single components (DKP12.5; DKP25; TRAM37.5; the active control ibuprofen blinded, there was also a
TRAM75). An active control (ibuprofen 400 mg, as an acid placebo tablet matching ibuprofen (each single-dose
formulation) was included in order to demonstrate the treatment consisted of two tablets) leading to a double-
sensitivity of the pain model, because it was significantly dummy design. The blind was maintained for patients and
superior to placebo in trials in the same indication [18–20], for people responsible for the ongoing conduct of the
and has the largest body of data for the indication [3]. study (such as the management, monitors, investigators)
The overall study duration was approximately 30 days and those responsible for data analysis and interpret-
for each patient, including three visits to the study site: ation of results at the conclusion of the study, such
Visit 1, for screening (within two weeks of their scheduled as biometrics personnel.
surgery day); Visit 2, for dental surgery, randomisation Rescue medication (RM) consisting of paracetamol 1 g
and treatment administration, followed by a 24-hour (with a maximum recommended daily dose of 4 g) was
post-dose pain and analgesic effect assessment period available on request during the 24-hour post-dose period.
(with the first four hours at the study site), during Patients were encouraged but not compelled to wait
which patients had to record efficacy data using an for at least 60 min post-dose, to allow time for the
electronic diary (eDiary); and Visit 3 (End of Study), study medication to take effect.
for final safety follow-up (10 ± 3 days after surgery day). In
addition, patients received a phone call for safety assess- Efficacy evaluation
ment the day after Visit 2 (approximately 24 h post-dose). Following treatment administration, patients were re-
For those patients who met the selection criteria, the quested to make multiple assessments of pain intensity
surgical procedure was performed under standardised and pain relief (PAR) on the eDiary over a period of 24 h.
local anaesthesia, which was limited to local anaesthetic They also had to make an overall assessment of the study
block using lidocaine (2 %) with epinephrine (1:80.000) medication (patient global evaluation, PGE) at the
up to a total volume of 5.4 mL per molar. After surgery, end of this period. The time when RM was first used,
patients reporting pain were asked to rate their pain if applicable, was also recorded.
intensity (PI) by a VAS (0–100; with the left end labelled PI was measured on a 4-point VRS (0 = none, 1 = mild,
“no pain” and the right end labelled “worst possible 2 = moderate, 3 = severe; [21]) immediately prior the
pain”; [21, 22]) and by a 4-point VRS (0 = none, 1 = mild, administration of study medication (baseline PI) and
2 = moderate, 3 = severe; [21]) to assess their eligibility then at 15 min, 30 min, 45 min, 1 h, 1.5 h, 2 h, 2.5 h,
for randomisation. Patients experiencing pain of moderate 3 h, 3.5 h, 4 h, 5 h, 6 h, 8 h, 12 h and 24 h post-dose. PAR
to severe intensity (VAS ≥40 mm and VRS ≥ 2) within four was measured on a 5-point VRS (0 = none, 1 = slight,
hours after the end of surgery were randomised and 2 = moderate, 3 = good, 4 = complete; [21]) at the
received one single oral dose of the assigned study same pre-defined post-dose time points. PGE was
treatment. After randomisation and immediately prior measured on a 5-point VRS (1 = poor, 2 = fair, 3 = good,
to the administration of study medication, VRS-PI was 4 = very good, 5 = excellent; [21, 23]) at 24 h post-dose (or
measured again and the score was recorded as baseline PI whenever the patient used RM, if this occurred first).
for the efficacy analysis. When patients used RM, a final PI and PAR assessment
Participants were randomly assigned to one of 10 and the PGE were recorded immediately before the intake
treatment groups following a blocked randomisation and, after that, they were excluded from further efficacy
procedure, with a block size of 10 and an allocation ratio measurements. After use of RM, the baseline observation
of 1:1:1:1:1:1:1:1:1:1. The randomisation process was carried forward (BOCF) method was applied [24], with PI
centralised by an Interactive Voice/Web Response returning to its baseline score and PAR to zero for all
System (IVRS/IWRS) and the treatment code was subsequent time points. If a patient prematurely withdrew
delivered for each patient according to a computer- from the study, final PI and PAR assessment and the PGE
generated random allocated sequence (randomisation list) were also requested.
prepared by a Sponsor’s third party prior to the start of From the PI and PAR scores, the summed pain intensity
the study. Two sets were prepared, one set was used for differences (SPID) and the total pain relief (TOTPAR)
programming the IVRS/IWRS and the other set was over 4, 6, 8 and 12 h post-dose were calculated. SPID was
used for the labelling of the study medication. Personnel calculated as the time-weighted sum of the pain intensity
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 4 of 13

difference (PID) values from baseline and TOTPAR was means of descriptive statistics; quantitative variables
calculated as the time-weighted sum of the PAR scores. (TOTPAR, % max TOTPAR, SPID and % max SPID)
The percentages of the theoretical maximum possible showing homogeneity of variance (according to Levene’s
SPID (% max SPID) and of the theoretical maximum test) were analysed by one-way analysis of variance
possible TOTPAR (% max TOTPAR) were also calculated. (ANOVA) using the Dunnett’s test for comparison between
The primary efficacy endpoint was the percentage of placebo and each active treatment, with an overall signifi-
patients showing response, defined as the achievement cance level of 5 % two-sided; ordinal variables (PGE) and
of at least 50 % of the maximum possible TOTPAR quantitative variables showing no homogeneity of variance
(≥50 % max TOTPAR), over 6 h post-dose within the were analysed by the Wilcoxon rank-sum test for
respective treatment arm [3, 24]. comparison between placebo and each active treatment,
Secondary efficacy endpoints included: percentage of using the Hochberg correction for the adjustment for
responders (≥50 % max TOTPAR) over 4, 8 and 12 h; multiple comparisons, with an overall significance level of
mean PI and PAR (VRS) scores over 24 h; SPID, % max 5 % two-sided; time to RM was analysed using the
SPID, TOTPAR and % max TOTPAR over 4, 6, 8 and 12 h; Kaplan-Meier estimation method and treatment groups
PGE at the end of the assessment period; time to first use were compared using a log-rank test, with the Hochberg
of RM since treatment administration and percentage of method applied for the multiplicity correction; percentage
patients using RM over 4, 6, 8, 12 and 24 h. of patients using RM were analysed analogously to the
primary efficacy variable. Safety variables were analysed by
Safety evaluation means of descriptive statistics.
The safety evaluation was based on the incidence, As an exploratory analysis, the primary efficacy endpoint
seriousness, intensity and causal relationship of treatment- was reassessed using active control as comparator. The f
emergent adverse events (AEs). AEs were assessed statistics test was used to evaluate if the effect of dexketo-
throughout the entire study by means of a non-leading profen, tramadol and their combination was statistically
open question. Spontaneously reported AEs were also significant on the outcome. When at least one of the
recorded. Furthermore, safety was also evaluated by the three f tests was found to be statistically significant,
assessment of clinically significant changes post-dose the Tukey method was applied to find out which
versus baseline in physical examination, vital signs doses gave a significant difference on the outcome.
(VS; blood pressure and heart rate), 12-lead electrocardio- Statistical differences between NNTs were examined
gram (ECG) and laboratory safety tests (haematology, using the z-test [25].
biochemistry and urinalysis). Any patient who prema- All efficacy analyses were performed on the “intention-
turely withdrew after having received study medication to-treat” (ITT) population (randomised patients who
was encouraged to undergo Visit 3. received study medication and for whom at least one
post-dose assessment was available). The “per protocol”
Statistical analysis (PP) population (patients of the ITT population with no
For the primary efficacy variable, in order to demonstrate major protocol violations) was used to perform confirma-
the superiority of active treatment in comparison with tory analyses on the primary endpoint. Safety analyses
placebo the null hypothesis of equality between placebo were performed on the “safety” population (randomised
and each tested study drug (the four combinations and patients who received study medication).
the four corresponding single agents) was tested using a It was estimated that 540 evaluable patients (60 per
Chi-square test. Multiplicity was adjusted by using the treatment arm) would have to be included in order to
Šidák correction [α =1-(1- α) ^ (1/k)], where k was the achieve approximately 80 % power in rejecting the null
number of comparisons. Considering eight comparisons, a hypothesis of equality between placebo and the eight
type I error probability = 0.00639 was used for the single experimental treatment arms (the four combinations
comparisons. The null hypothesis of equality between and the four corresponding single agents) regarding
placebo and the active control was also tested to validate the primary endpoint, on the basis of the following
the pain model. In addition, event rates (ER), relative risk assumptions [26]: response rate for placebo = 0.13; expected
(RR), NNT and relative risk reduction (RRR), with RR of response in active treatment versus placebo = 3.21;
their corresponding 95 % confidence intervals (CI), overall type I error probability of 0.05 (two-sided). Sixty
were estimated in order to compare the effect size of further patients were to be treated with the active control
placebo with the effect size of each active treatment. as a reference for the effect of the eight experimental
Secondary efficacy variables were analysed as follows: treatment arms. It was expected that approximately 667
percentage of responders over 4, 8 and 12 h were patients would have to be screened in order to obtain 600
analysed analogously to the primary efficacy variable; randomised patients, assuming an approximate rate of
mean PI and PAR (VRS) scores were analysed by 10 % screening failures.
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 5 of 13

Results superior for DKP25/TRAM75, DKP12.5/TRAM75, DKP25/

Of the 745 patients screened, 611 patients were ran- TRAM37.5 (p < 0.0001) and DKP25 (p = 0.0012) over eight
domised and received the study treatment, thus con- hours; and for DKP25/TRAM75 (p = 0.0002), DKP12.5/
stituting the safety population. Efficacy analyses were TRAM75 (p = 0.0004) and DKP25/TRAM37.5 (p = 0.0028)
performed on the ITT population of 606 randomised over 12 h. The highest percentage of responders over
patients. The PP population of 567 patients was used four, eight and 12 h was achieved with DKP25/TRAM75
to perform confirmatory analyses on the primary end- (79 %, 54 % and 38 % respectively) versus 6.5 % with
point. Patient assignment to the different populations placebo. Results are represented in Additional file 4
occurred before the study blind was broken. The partici- and Additional file 5.
pant flow with the numbers of participants who were The active control, ibuprofen 400 mg, was statistically
randomly assigned, received intended treatment, and superior to placebo (p < 0.0001), thus validating the pain
were analysed for the primary outcome is represented model. Analyses run on the PP population confirmed
in Additional file 1. the primary efficacy results.
Demography and baseline characteristics of different
treatment groups were comparable. Demographic and Secondary endpoints
baseline characteristics of the ITT population are repre- The time course of mean PAR and PI over the whole
sented in Table 1. The overall mean age was 27 years 24 h post-dose demonstrated rapid onset of pain relief
(range 18–64 years), 59 % were women, and 90 % were with dexketoprofen alone or in combination, and that
white. The mean surgery time was 29 min, most patients the addition of tramadol to dexketoprofen resulted both
(90 %) had only one mandibular molar removed, and the in greater peak pain relief and greater pain relief over
total number of molar extractions was one or two in the longer term, particularly at times longer than six
95 % of patients. Initial pain (baseline PI) was moderate hours post dose (Fig. 3, and Additional file 6).
or severe in 601 patients (Table 2). The analysis of summary efficacy measures (SPID, % max
Four (0.65 %) patients, out of 611 randomised, discon- SPID, TOTPAR and % max TOTPAR; Additional file 7,
tinued the study after randomisation, thus resulting in a Additional file 8, Additional file 9, Additional file 10 and
total of 607 patients completing the study. One patient Additional file 11) showed that all DKP/TRAM com-
(allocated to DKP12.5/TRAM75) discontinued the binations and both doses of DKP in monotherapy
study due to “failure of eDiary”. The other three patients were significantly superior to placebo (p < 0.01) except for
(allocated respectively to DKP25/TRAM37.5, DKP25/ DKP12.5 over 12 h, with the best results achieved with
TRAM75 and TRAM37.5) were “lost to follow-up”. Three DKP25/TRAM75. The time course of mean SPID and
of these four patients attended Visit 3. mean TOTPAR are represented in Additional file 12 and
Additional file 13.
Efficacy results The time to RM was significantly longer (p < 0.005) for
Primary endpoint all active treatments (except for both doses of TRAM.HCl
The percentage of patients with ≥ 50 % max TOTPAR in monotherapy) than for placebo, with DKP12.5/TRAM75
over six hours post-dose was significantly superior to pla- and DKP25/TRAM75 presenting the longest value
cebo for all DKP/TRAM combinations and also for (median time, [95 % CI]: 8.5 h [5.9 to 13.0] and 8.1 h
DKP25 (p < 0.0001 for each comparison, except p = 0.0009 [6.3 to 13.4] respectively, versus 1.4 h [1.2 to 1.8] in the
for DKP12.5/TRAM37.5), with the highest percentage placebo group). Fig. 4 shows the proportion of patients
of responders achieved in the DKP25/TRAM75 group remedicating over time in each group, with additional
(72 % versus 10 % in the placebo group; Fig. 1 and information in Additional file 14 and Additional file 15.
Additional file 2. The percentage of patients requiring RM (Additional
Detailed ER, RR and NNT results at six hours are file 16 and Additional file 17) over four and six hours was
presented in Additional file 3. DKP25/TRAM75 also significantly smaller (p < 0.00639) for DKP12.5/TRAM75,
had the highest RR [7.2 (95 % CI: 3.3 to 15.7)] and DKP25/TRAM37.5, and DKP25/TRAM75 (also for DKP25
the lowest NNT [1.6 (95 % CI: 1.3 to 2.1)]. No other over four hours) than for placebo (over six hours 47 %,
point estimate for NNT was below 2.0 (Fig. 2). The 40 %, and 38 % for each combination respectively versus
NNT for DKP25/TRAM75 was significantly better than 73 % for placebo). The difference remained significant for
other NNT values (p < 0.05, z-test), except for DKP25/ DKP25/TRAM75 over eight hours (48 % versus 73 %) and
TRAM37.5, DKP12.5/TRAM75 and DKP25. for DKP25/TRAM37.5 over 12 and 24 h (49 % versus 74 %
The percentage of responders (≥50 % max TOTPAR) for both time points).
was significantly superior to placebo for all DKP/TRAM The PGE of the study medication at the end of the
combinations and for both doses of DKP in monotherapy assessment period is represented in Fig. 5. All treat-
(p < 0.0001) over four hours and it remained significantly ments (except TRAM37.5) were significantly superior
 Moore et al. The Journal of Headache and Pain (2015) 16:60
Table 1 Demographic and baseline characteristics (ITT population)
DKP 12.5 mg + DKP 12.5 mg + DKP 25 mg + DKP 25 mg + DKP DKP 25 mg TRAM 37.5 mg TRAM 75 mg Ibuprofen Placebo Overall
TRAM 37.5 mg TRAM 75 mg TRAM 37.5 mg TRAM 75 mg 12.5 mg
n 60 62 63 61 60 60 59 59 60 62 606
Gender n (%) female 34 (56.7) 38 (61.3) 36 (57.1) 34 (55.7) 36 (60.0) 43 (71.7) 38 (64.4) 27 (45.8) 40 (66.7) 33 (53.2) 359 (59.2)
male 26 (43.3) 24 (38.7) 27 (42.9) 27 (44.3) 24 (40.0) 17 (28.3) 21 (35.6) 32 (54.2) 20 (33.3) 29 (46.8) 247 (40.8)
Ethnic origin n (%) White 56 (93.3) 57 (91.9) 56 (88.9) 59 (96.7) 54 (90.0) 51 (85.0) 56 (94.9) 52 (88.1) 49 (81.7) 57 (91.9) 547 (90.3)
Asian 2 (3.3) 2 (3.2) 5 (7.9) 2 (3.3) 5 (8.3) 6 (10.0) 3 (5.1) 2 (3.4) 5 (8.3) 2 (3.2) 34 (5.6)
Black 2 (3.3) 2 (3.2) 1 (1.6) 0 (0.0) 1 (1.7) 2 (3.3) 0 (0.0) 4 (6.8) 5 (8.3) 2 (3.2) 19 (3.1)
Other 0 (0.0) 1 (1.6) 1 (1.6) 0 (0.0) 0 (0.0) 1 (1.7) 0 (0.0) 1 (1.7) 1 (1.7) 1 (1.6) 6 (1.0)
Age (years) mean (SD) 28.7 (7.71) 27.0 (7.66) 26.3 (7.33) 27.3 (7.55) 27.0 (9.85) 27.0 (6.94) 25.5 (7.15) 27.9 (8.04) 26.7 (6.48) 26.1 (6.64) 26.9 (7.57)
range 18–52 18–53 18–64 18–52 18–63 18–48 18–55 18–58 18–44 18–54 18–64
BMI (kg/m2) mean (SD) 23.7 (3.38)a 24.1 (3.69) 23.0 (2.87) 23.2 (3.19) 23.6 (3.20) 23.6 (3.30)b 23.0 (3.19) 24.2 (3.10) 22.4 (3.06) 22.7 (2.80) 23.3 (3.21)
range 18–30 18–35 18–32 18–30 18–30 15–30 18–30 18–34 18–31 18–29 18–35
Surgery durationc mean (SD) 29:02 (14:40) 28:45 (12:58) 29:35 (17:32) 30:57 (16:39) 29:01 (12:03) 29:39 (14:45) 27:10 (11:10) 27:37 (14:01) 29:56 (13:31) 32:28 (15:23) 29:26 (14:23)
Total third molar 1 38 (63.3) 40 (64.5) 38 (60.3) 38 (62.3) 41 (68.3) 38 (63.3) 36 (61.0) 35 (59.3) 31 (51.7) 35 (56.5) 370 (61.1)
extractionsd n (%)
2 17 (28.3) 20 (32.3) 23 (36.5) 18 (29.5) 17 (28.3) 20 (33.3) 22 (37.3) 21 (35.6) 26 (43.3) 22 (35.5) 206 (34.0)
3 3 (5.0) 0 (0.0) 0 (0.0) 2 (3.3) 0 (0.0) 1 (1.7) 1 (1.7 ) 2 (3.4) 2 (3.3) 3 (4.8) 14 (2.3)
4 2 (3.3) 2 (3.2) 2 (3.2) 3 (4.9) 2 (3.3) 1 (1.7) 0 (0.0) 1 (1.7) 1 (1.7) 2 (3.2) 16 (2.6)
Lower third molar 1 54 (90.0) 55 (88.7) 56 (88.9) 53 (86.9) 55 (91.7) 53 (88.3) 56 (94.9) 52 (88.1) 53 (88.3) 57 (91.9) 544 (89.8)
extractions n (%)
2 6 (10.0) 7 (11.3) 7 (11.1) 8 (13.1) 5 (8.3) 7 (11.7) 3 (5.1) 7 (11.9) 7 (11.7) 5 (8.1) 62 (10.2)
BMI body mass index
a
n = 59; bn = 59; ctime is expressed in minutes and seconds; dtotal number of third molar extractions, including also upper third molar teeth

Page 6 of 13
 Moore et al. The Journal of Headache and Pain (2015) 16:60
Table 2 PI before randomization and before treatment administration (ITT population)
DKP 12.5 mg + DKP 12.5 mg + DKP 25 mg + DKP 25 mg + DKP 12.5 mg DKP 25 mg TRAM 37.5 mg TRAM 75 mg Ibuprofen Placebo Overall
TRAM 37.5 mg TRAM 75 mg TRAM 37.5 mg TRAM 75 mg
n 60 62 63 61 60 60 59 59 60 62 606
PI before randomization
VAS Mean (SD) 58.28 (12.50) 57.74 (11.54) 56.25 (10.79) 57.72 (11.63) 58.33 (12:49) 57.72 (12.55) 57.10 (11:73) 56.32 (11.42) 57.33 (13:46) 60.34 (13.39) 57.72 (12.13)
VRS Moderate n (%) 48 (80.0) 53 (85.5) 48 (76.2) 48 (78.7) 45 (75.0) 47 (78.3) 56 (94.9) 47 (79.7) 51 (85.0) 46 (74.2) 489 (80.7)
Severe n (%) 12 (20.0) 9 (14.5) 15 (23.8) 13 (21.3) 15 (25.0) 13 (21.7) 3 (5.1) 12 (20.3) 9 (15.0) 16 (25.8) 117 (19.3)
PI before treatment administration (baseline PI)
VRS Mild n (%) 1 (1.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (1.7) 0 (0.0) 1 (1.6) 3 (0.5)
Moderate n (%) 40 (66.7) 43 (69.4) 39 (61.9) 41 (67.2) 35 (58.3) 43 (71.7) 40 (67.8) 35 (59.3) 39 (65.0) 33 (53.2) 388 (64.0)
Severe n (%) 19 (31.7) 18 (29.0) 24 (38.1) 20 (32.8) 25 (41.7) 17 (28.3) 19 (32.2) 23 (39.0) 21 (35.0) 27 (43.5) 213 (35.1)
Missing n (%) 0 (0.0) 1 (1.6) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (1.6) 2 (0.3)
VRS-PI measured on a 4-point VRS (0 = ‘none’ to 3 = ‘severe’). Baseline PI refers to the VRS-PI recorded immediately prior to the administration of the study medication (in contrast to VRS-PI measured before
randomisation for eligibility purposes). Baseline PI was moderate or severe in 601 patients; it was reported as “mild” by 3 (0.5 %) patients and results were missing for 2 (0.3 %) patients due to compilation data error
on the eDiary

Page 7 of 13
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 8 of 13

Fig. 1 Percentage of patients showing response (≥50 % max TOTPAR) over 6 h post-dose (Primary Endpoint). Maximum TOTPAR corresponds to
the theoretical maximum possible time-weighted sum of the PAR scores, measured on a 5-point VRS (0 = ‘none’ to 4 = ‘complete’)

to placebo (p < 0.01), with the highest scores in the vomiting (21 patients; 3.4 %), nausea (14 patients; 2.3 %),
DKP25/TRAM75 group. The percentage of patients with dizziness (11 patients; 1.8 %) and somnolence (5 patients;
‘good’ to ‘excellent’ PGE response was 79 % in the DKP25/ 0.8 %) (Table 3). Apart from one case of “severe” somno-
TRAM75 group versus 11 % in the placebo group. The lence in the DKP12.5/TRAM75 group (1.6 %), all ADRs
percentage of patients with ‘very good’ and ‘excellent’ PGE were considered “mild” (45 ADRs, 71 %) or “moderate”
was 51 % in the DKP25/TRAM75 group versus 4.8 % in (17 ADRs, 27 %) in intensity.
the placebo group. Only one serious adverse event (SAE) was reported in
The analysis of dose–response relationship between the one patient (allocated to the TRAM75 group), consisting in
tested study drugs and the active control showed that dizziness of mild intensity, which required hospitalization
DKP25/TRAM75 was the only combination that was for monitoring and resolved spontaneously. The event was
significantly superior to ibuprofen 400 mg (p = 0.0028). assessed as “possibly related” to the study medication.
Dizziness is a commonly reported ADR associated to
Safety results the use of tramadol, occurring in more than 10 % of
Of 611 patients treated, 40 (6.5 %) reported a total of patients, according to the authorized Summary of Product
63 adverse reactions (ADRs), the most frequent being Characteristics.

Fig. 2 NNT for ≥ 50 % max TOTPAR compared with placebo over six hours post dose. Maximum TOTPAR corresponds to the theoretical
maximum possible time-weighted sum of the PAR scores, measured on a 5-point VRS (0 = ‘none’ to 4 = ‘complete’). Bars show 95 % confidence
interval of NNT, with colour change as point estimate (Note that TRAM37.5 was not significantly better than placebo)
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 9 of 13

Fig. 3 Time course of mean PAR scores (0–24 h). PAR measured on a 5-point VRS (0 = ‘none’ to 4 = ‘complete’)

The highest incidence of ADRs was reported in the No deaths or other significant AEs occurred. No
TRAM75 group (17 %). This incidence was higher patient discontinued because of AEs. There were no
than reported in groups receiving tramadol 75 mg in clinically relevant changes in the VS, physical examination,
combination (i.e. DKP12.5/TRAM75 and DKP25/ 12-lead ECG or laboratory safety tests versus baseline.
TRAM75) for which the incidence was of 9.5 % and Overall, all treatments were safe and well tolerated, with
12 % of patients, respectively. all DKP/TRAM combinations presenting a safety and

Fig. 4 Cumulative frequency (Kaplan-Meier estimation) of RM intake (0–24 h)

Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 10 of 13

Fig. 5 PGE at the end of the assessment period. PGE measured on a 5-point VRS (1 = ‘poor’ to 5 = ‘excellent’)

tolerability profile fully in line with that previously estab- extrapolation to clinical practice should not be under-
lished for the single agents. taken. However, the NNT for ibuprofen 400 mg was mea-
sured in this trial as 2.9 (95 % confidence interval 2.0 to
Discussion 4.9); this is comfortably close to that of 2.5 (2.4 to 2.6)
The results of this trial demonstrate good results from a found in a meta-analysis of over 6000 patients [20],
combination of a rapidly-acting NSAID, dexketoprofen, and that for dexketoprofen trometamol 25 mg of 2.2
with a longer lasting opioid with additional enhance- (1.7 to 3.3) close to that of 3.2 (2.6 to 4.1) in another
ment of serotonin and norepinephrine transmission [27]. meta-analysis [13], or 2.6 (2.0 to 3.5) in dental pain
The particular combination of dexketoprofen trometa- in an analysis from clinical trial reports [14]. Information
mol 25 mg plus tramadol hydrochloride 75 mg delivered on tramadol hydrochloride 75 mg in single dose studies is
rapid initial pain relief, low (good) NNT for at least 50 % limited and variable [16, 17], but the wide confidence
max TOTPAR over six hours, long duration, and a small interval around the measured NNT of 6.7 (3.5 to 60)
proportion of patients remedicating. The NNT of 1.6 and encompasses existing estimates.
the eight hours before 50 % of patients remedicated were The other interesting, and perhaps important, result of
comparable to or better than most other oral treatments this trial was that the pre-trial estimates of efficacy of
for acute postoperative pain [3]. the combinations, determined using a formula derived
This makes DKP25/TRAM75 a good candidate for from empirical data, proved to be reasonably accurate
a combination treatment in acute pain. In part, this is predictors of measured efficacy. Because of limited
because the combination unites two commonly used efficacy data for single dose dexketoprofen and tramadol
drugs (dexketoprofen is the active chiral form of at the doses envisaged, the pre-trial estimates could
ketoprofen) with known properties. This is important be little more than informed guesses, but they were
because new rare but serious adverse events are less likely not inaccurate.
than, say, with a new chemical. But efficacy, tolerability,
and safety need to be tested in more trials and more Conclusion
patients, perhaps especially in multiple dose studies Dexketoprofen trometamol 25 mg combined with tramadol
that capture more information than a single dose efficacy hydrochloride 75 mg provided good all-round analgesia,
study can do. with rapid onset and long duration in a model of moderate
Although this trial was performed to high quality to severe pain. The results of the dose finding study are
standards, and was relatively large with over 600 patients consistent with pre-trial calculations based on empirical
recruited, the modest size of each group means that formulae.
 Moore et al. The Journal of Headache and Pain (2015) 16:60
Table 3 ADRs - by system organ class /preferred term, by treatment group and overall
System organ class Preferred term DKP 12.5 mg + DKP 12.5 mg + DKP 25 mg + DKP 25 mg + DKP 12.5 mg DKP 25 mg TRAM 37.5 mg TRAM 75 mg Ibuprofen Placebo Overall
TRAM 37.5 mg TRAM 75 mg TRAM 37.5 mg TRAM 75 mg
n = 61 n = 63 n = 63 n = 61 n = 60 n = 61 n = 59 n = 60 n = 61 n = 62 n = 611
Gastrointestinal 3 (4.9)|5 4 (6.3)|4 3 (4.8)|3 6 (9.8)|8 1 (1.7)|1 3 (4.9)|3 0 8 (13.3)|13 1 (1.6)|1 0 29 (4.7)|38
disorders
Abdominal pain 0 0 0 0 0 0 0 1 (1.7)|1 0 0 1 (0.2)|1
upper
Nausea 3 (4.9)|3 0 0 3 (4.9)|3 1 (1.7)|1 1 (1.6)|1 0 6 (10.0)|7 0 0 14 (2.3)|15
Vomiting 2 (3.3)|2 4 (6.3)|4 3 (4.8)|3 4 (6.6)|5 0 2 (3.3)|2 0 5 (8.3)|5 1 (1.6)|1 0 21 (3.4)|22
General disorders 0 1 (1.6)|1 0 2 (3.3)|2 0 1 (1.6)|1 1 (1.7)|1 0 1 (1.6)|1 0 6 (1.0)|6
and administration
site conditions Chills 0 1 (1.6)|1 0 0 0 0 0 0 0 0 1 (0.2)|1
Discomfort 0 0 0 1 (1.6)|1 0 0 0 0 1 (1.6)|1 0 2 (0.3)|2
Feeling abnormal 0 0 0 1 (1.6)|1 0 0 0 0 0 0 1 (0.2)|1
Pyrexia 0 0 0 0 0 1 (1.6)|1 1 (1.7)|1 0 0 0 2 (0.3)|2
Nervous System 1 (1.6)|1 4 (6.3)|4 1 (1.6)|1 3 (4.9)|3 0 0 2 (3.4)|2 3 (5.0)|4 2 (3.3)|2 0 16 (2.6)|17
disorders
Dizziness 1 (1.6)|1 2 (3.2)|2 0 2 (3.3)|2 0 0 1 (1.7)|1 3 (5.0)|3 2 (3.3)|2 0 11 (1.8)|11
Headache 0 0 0 0 0 0 1 (1.7)|1 0 0 0 1 (0.2)|1
Somnolence 0 2 (3.2)|2 1 (1.6)|1 1 (1.6)|1 0 0 0 1 (1.7)|1 0 0 5 (0.8)|5
Psychiatric disorders 0 0 0 0 0 0 0 1 (1.7)|1 0 0 1 (0.2)|1
Nervousness 0 0 0 0 0 0 0 1 (1.7)|1 0 0 1 (0.2)|1
Vascular disorders 0 0 0 0 0 0 0 1 (1.7)|1 0 0 1 (0.2)|1
Hypotension 0 0 0 0 0 0 0 1 (1.7)|1 0 0 1 (0.2)|1
Overall 3 (4.9)|6 6 (9.5)|9 4 (6.3)|4 7 (11.5)|13 1 (1.7)|1 3 (4.9)|4 3 (5.1)|3 10 (16.7)|19 3 (4.9)|4 0 40 (6.5)|63
Results are expressed as number of patients (% of exposed) | number of events

Page 11 of 13
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 12 of 13

Additional files Gesichtschirurgie, Plastische Operationen, Kiel, Germany), HR Metelmann

(Universitätsmedizin Greifswald, Klinik und Poliklinik für Mund-Kiefer-
Gesichtschirurgie / Plastische Operationen, Greifswald, Germany), R Baca
Additional file 1: Study CONSORT flow diagram. Participant flow with
Pérez-Bryan (Universidad Complutense de Madrid, Facultad de Odontología,
the numbers of participants who were randomly assigned, received
Madrid, Spain), and to the study data management from A Madami (Menar-
intended treatment, and were analysed for the primary outcome.
ini Ricerche S.p.A.).
Additional file 2: Cumulative percentage of patients showing
response (≥50 % max TOTPAR) over 6 h post-dose (Primary Author details
Endpoint). Maximum TOTPAR corresponds to the theoretical maximum 1
Pain Research and Nuffield Division of Anaesthetics, Nuffield Department of
possible time-weighted sum of the PAR scores, measured on a 5-point Clinical Neurology, University of Oxford, The Churchill, Oxford OX3 7LE, UK.
VRS (0 = ‘none’ to 4 = ‘complete’). 2
Department of Oral and Maxillofacial Surgery, School of Dentistry, University
Additional file 3: Effect size on the percentage of patients showing of Barcelona, Bellvitge Institute for Biomedical Research (IDIBELL), Barcelona,
response (≥50 % max TOTPAR) over 6 h - Estimated ER, RR and NNT Spain. 3Dr. Tóth Bagi Zoltán Fogászati Rendeloje, Budapest, Hungary. 4The
(95 % CI). School of Dentistry, College of Medical and Dental Sciences, University of
Birmingham, Birmingham, UK. 5Centro Ricerche Cliniche, Ospedale Policlinico
Additional file 4: Percentage of patients showing response
G.B.Rossi, Verona, Italy. 6Hospital Universitario Virgen del Rocío,
(≥50 % max TOTPAR) over 4 h, 6 h (Primary Endpoint), 8 h and
Departamento de Estomatología, Facultad de Odontología, Universidad de
12 h post-dose.
Sevilla, Sevilla, Spain. 7University Dental Hospital, Cardiff, UK. 8Department of
Additional file 5: Statistical Analysis of the percentage of patients Maxillofacial Surgery, Hospital Médico Quirúrgico de Conxo, Santiago de
showing response (≥50 % max TOTPAR) over 4 h, 6 h (Primary Compostela, Spain. 9School of Dentistry, University of Manchester,
Endpoint), 8 h and 12 h post-dose. Manchester, UK. 10Department of Oral Surgery, Warsaw Medical Univeristy,
Additional file 6: Time course of mean VRS-PI scores (0–24 h). Warsaw, Poland. 11Departament d’Estomatologia, Clínica Odontológica,
VRS-PI measured on a 4-point VRS (0 = ‘none’ to 3 = ‘severe’). Universidad de Valencia, Valencia, Spain. 12Clinical Research, Laboratorios
Menarini S.A. – Menarini Group, Badalona, Spain. 13Clinical Research, Menarini
Additional file 7: Descriptive statistics (mean and SD) for SPID,
Ricerche S.p.A – Menarini Group, Florence, Italy.
% max SPID, TOTPAR and % max TOTPAR over 4, 6, 8 and 12 h.
Additional file 8: Statistical Analysis of TOTPAR over 4, 6, 8 and Received: 29 April 2015 Accepted: 12 June 2015
12 h.
Additional file 9: Statistical analysis of percentage of max TOTPAR
over 4, 6, 8 and 12 h.
Additional file 10: Statistical analysis of SPID over 4, 6, 8 and 12 h. References
1. Moertel CG, Ahmann DL, Taylor WF, Schwartau N (1974) Relief of pain by
Additional file 11: Statistical analysis of percentage of max SPID
oral medications. A controlled evaluation of analgesic combinations.
over 4, 6, 8 and 12 h.
JAMA 229(1):55–9
Additional file 12: Time course of mean SPID (0–12 h). PI measured 2. McQuay HJ, Derry S, Eccleston C, Wiffen PJ, Moore RA (2012) Evidence for
on a 4-point VRS (0 = ‘none’ to 3 = ‘severe’); PID = PI t0h (baseline PI) – PI t analgesic effect in acute pain - 50 years on. Pain 153(7):1364–7
(PI at time point t). (TIFF 86 kb) 3. Moore RA, Derry S, McQuay HJ, Wiffen PJ (2011) Single dose oral analgesics
Additional file 13: Time course of mean TOTPAR (0–12 h). PAR for acute postoperative pain in adults. Cochrane Database Syst Rev
measured on a 5-point VRS (0 = ‘none’ to 4 = ‘complete’). 7(9):CD008659
4. Toms L, Derry S, Moore RA, McQuay HJ (2009) Single dose oral paracetamol
Additional file 14: Kaplan-Meier survival distribution of the time to
(acetaminophen) with codeine for postoperative pain in adults.
RM (0–24 h). Time to RM is defined as the time elapsed between the
Cochrane Database Syst Rev 21(1):CD001547
treatment administration and the first RM use; the survival distribution
5. Gaskell H, Derry S, Moore RA, McQuay HJ (2009) Single dose oral oxycodone
estimates the probability of ‘no use of RM’ at each time point.
and oxycodone plus paracetamol (acetaminophen) for acute postoperative
Additional file 15: Time to RM (Kaplan-Meier estimation) over 24 h. pain in adults. Cochrane Database Syst Rev 8(3):CD002763
Additional file 16: Cumulative percentage of patients who required 6. Derry S, Karlin SM, Moore RA (2013) Single dose oral ibuprofen plus codeine
RM over 4, 6, 8, 12 and 24 h. for acute postoperative pain in adults. Cochrane Database Syst Rev
Additional file 17: Statistical analysis of percentage of patients who 28(3):CD010107
required RM over 4, 6, 8, 12 and 24 h. 7. Derry S, Derry CJ, Moore RA (2013) Single dose oral ibuprofen plus
oxycodone for acute postoperative pain in adults. Cochrane Database Syst
Rev 26(6):CD010289
Competing interests 8. Derry CJ, Derry S, Moore RA (2013) Single dose oral ibuprofen plus
The study was funded by Menarini Group. RAM and CGE have been paid paracetamol (acetaminophen) for acute postoperative pain. Cochrane
consultants for Menarini Group and have been members of speakers Database Syst Rev 24(6):CD010210
bureaux, but received no direct funding for this work. SC, BPV, CGB, MB, 9. Derry CJ, Derry S, Moore RA (2012) Caffeine as an analgesic adjuvant for
MPC, SS, AN, AC and CAM are employees of Menarini Group. The other acute pain in adults. Cochrane Database Syst Rev 14(3):CD009281
authors report no competing interests. 10. Moore RA, Derry CJ, Derry S, Straube S, McQuay HJ (2012) A conservative
method of testing whether combination analgesics produce additive or
Authors’ contributions synergistic effects using evidence from acute pain and migraine. Eur J Pain
The design of study was performed by RAM, CGE, SC, BPV, CGB, MB, MPC, 16(4):585–91
AC and CAM. The study was conducted by CGE, RF, ZTB, TD, SM, DTL, CMH, 11. Moore RA, Derry S, Straube S, Ireson-Paine J, Wiffen PJ (2014) Faster, higher,
AGG, PC, AW, DM and MPD. Data was analysed by RAM, SS, AN, SC, and BPV. stronger? Evidence for formulation and efficacy for ibuprofen in acute pain.
The manuscript was prepared by RAM, SC and BPV. All authors contributed Pain 155(1):14–21
to the development of interim and final drafts, and read and approved the 12. Moore RA, Derry S, Straube S, Ireson-Paine J, Wiffen PJ (2014) Validating
final manuscript. speed of onset as a key component of good analgesic response in acute
pain. Eur J Pain. 22. doi:10.1002/ejp.536.
Acknowledgements 13. Barden J, Derry S, McQuay HJ, Moore RA (2009) Single dose oral ketoprofen
The authors acknowledge the contribution to the conduct of the study and dexketoprofen for acute postoperative pain in adults. Cochrane Database
from L Bazzichi (Università degli Studi di Pisa, Division of Rheumatology, Syst Rev 7(4):CD007355
Department of Internal Medicine, Pisa, Italy), J Wiltfang (Universitätsklinikum 14. Moore RA, Barden J (2008) Systematic review of dexketoprofen in acute and
Schleswig-Holstein, Campus Kiel, Klinik für Mund-, Kiefer- und chronic pain. BMC Clin Pharmacol 8:11
Moore et al. The Journal of Headache and Pain (2015) 16:60 Page 13 of 13

15. Sawaddiruk P (2011) Tramadol hydrochloride/acetaminophen combination

for the relief of acute pain. Drugs Today (Barc) 47(10):763–72
16. Moore RA, McQuay HJ (1997) Single-patient data meta-analysis of 3453
postoperative patients: oral tramadol versus placebo, codeine and
combination analgesics. Pain 69(3):287–94
17. Edwards JE, McQuay HJ, Moore RA (2002) Combination analgesic efficacy:
individual patient data meta-analysis of single-dose oral tramadol plus
acetaminophen in acute postoperative pain. J Pain Symptom Manage
23(2):121–30
18. Cooper SA, Engel J, Ladov M, Precheur H, Rosenheck A, Rauch D (1982)
Analgesic efficacy of an ibuprofen-codeine combination. Pharmacotherapy
2(3):162–7
19. Mehlisch DR, Sollecito WA, Helfrick JF, Leibold DG, Markowitz R, Schow CE Jr,
Shultz R, Waite DE (1990) Multicenter clinical trial of ibuprofen and
acetaminophen in the treatment of postoperative dental pain. J Am
Dent Assoc 121(2):257–63
20. Derry C, Derry S, Moore RA, McQuay HJ (2009) Single dose oral ibuprofen
for acute postoperative pain in adults. Cochrane Database Syst Rev
8(3):CD001548
21. Barden J, Edwards JE, Mc Quay HJ, Moore RA (2004) Pain and analgesic
response after third molar extraction and other postsurgical pain.
Pain 107:86–90
22. Jamison RN, Gracely RH, Raymond SA, Levine JG, Marino B, Herrmann TJ,
Daly M, Fram D, Katz NP (2002) Comparative study of electronic vs. paper
VAS rating: a randomized, crossover trial using healthy volunteers.
Pain 99:341–7
23. Collins SL, Edwards J, Moore RA, Smith LA, Mc Quay HJ (2001) Seeking a
simple measure of analgesia for mega-trials: is a single global assessment
good enough? Pain 91:189–94
24. Moore RA, Edwards JE, McQuay HJ (2005) Acute pain: individual patient
meta-analysis shows the impact of different ways of analysing and
presenting results. Pain 116(3):322–31
25. Tramèr MR, Reynolds DJ, Moore RA, McQuay HJ (1997) Impact of covert
duplicate publication on meta-analysis: a case study. BMJ 315:635–640
26. Barden J, Edwards JE, Mc Quay HJ, Wiffen PJ, Moore RA (2004) Relative
efficacy of oral analgesics after third molar extraction. Br Dent J 197:407–11
27. Reeves RR, Burke RS (2008) Tramadol: basic pharmacology and emerging
concepts. Drugs Today (Barc) 44(11):827–36

Submit your manuscript to a

journal and benefit from:
7 Convenient online submission
7 Rigorous peer review
7 Immediate publication on acceptance
7 Open access: articles freely available online
7 High visibility within the field
7 Retaining the copyright to your article

Submit your next manuscript at 7 springeropen.com