RESEARCH ARTICLE Open Access
Dexke op o en/ amadol: andomised
double-blind ial and con i ma ion o empi ical
heo y o combina ion analgesics in acu e pain
R. And ew Moo e
1*
, C. Gay-Escoda
2
, R. Figuei edo
2
, Z. Tó h-Bagi
3
, T. Die ich
4
, S. Mille i
5
, D. To es-Laga es
6
,
C. M. Hill
7
, A. Ga cía-Ga cía
8
, P. Coul ha d
9
, A. Woj owicz
10
, D. Ma enko
10
, M. Peña ocha-Diago
11
, S. Cuad ipani
12
,
B. Pizà-Vallespi
12
, C. Gue e o-Bayón
12
, M. Be olo i
13
, M. P. Con ini
13
, S. Sca oni
13
, A. Nizza do
13
, A. Cap ia i
13
and C. A. Maggi
13
Abs ac
Backg ound: Combina ion analgesics a e e ec i e in acu e pain, and a heo e ical amewo k p edic s e icacy o
combina ions. The combina ion o dexke op o en and amadol is un es ed, bu p edic ed o be highly e ec i e.
Me hods: This was a andomised, double-blind, double-dummy, pa allel-g oup, placebo-con olled, single-dose ial
in pa ien s wi h mode a e o se e e pain ollowing hi d mola ex ac ion. The e we e en ea men a ms, including
dexke op o en ome amol (12.5 mg and 25 mg) and amadol hyd ochlo ide (37.5 mg and 75 mg), gi en as ou
di e en ixed combina ions and single componen s, wi h ibup o en 400 mg as ac i e con ol as well as a placebo
con ol. The s udy objec i e was o e alua e he supe io analgesic e icacy and sa e y o each combina ion and
each single agen e sus placebo. The p ima y ou come was he p opo ion o pa ien s wi h a leas 50 % max
TOTPAR o e six hou s.
Resul s: 606 pa ien s we e andomised and p o ided a leas one pos -dose assessmen . All combina ions we e
signi ican ly be e han placebo. The highes pe cen age o esponde s (72 %) was achie ed in he dexke op o en
ome amol 25 mg plus amadol hyd ochlo ide 75 mg g oup (NNT 1.6, 95 % con idence in e al 1.3 o 2.1).
Addi ion o amadol o dexke op o en esul ed in g ea e peak pain elie and g ea e pain elie o e he longe
e m, pa icula ly a imes longe han six hou s (median du a ion o 8.1 h). Ad e se e en s we e un ema kable.
Conclusions: Dexke op o en ome amol 25 mg combined wi h amadol hyd ochlo ide 75 mg p o ided good
analgesia wi h apid onse and long du a ion in a model o mode a e o se e e pain. The esul s o he dose
inding s udy a e consis en wi h p e- ial calcula ions based on empi ical o mulae.
T ial egis a ion: Eud aCT (2010-022798-32); Clinical ials.go (NCT01307020).
Keywo ds: Dexke op o en; T amadol; Combina ion analgesics; Pos ope a i e pain; Thi d mola ; Randomised
con olled ial; Dose ange
* Co espondence: [email p o ec ed]
1
Pain Resea ch and Nu ield Di ision o Anaes he ics, Nu ield Depa men o
Clinical Neu ology, Uni e si y o Ox o d, The Chu chill, Ox o d OX3 7LE, UK
Full lis o au ho in o ma ion is a ailable a he end o he a icle
© 2015 Moo e e al. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License
(h p://c ea i ecommons.o g/licenses/by/4.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium,
p o ided he o iginal wo k is p ope ly c edi ed.
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60
DOI 10.1186/s10194-015-0541-5
Backg ound
G ea e e icacy om combina ion analgesics in acu e
pain has been ecognised o some ime [1], albei
o iginally in cance pain. When o al analgesic d ugs
a e es ed in s anda d, acu e pain models [2] hose
wi h he highes e icacy and lowes numbe s-needed-
o- ea (NNT) a e ypically high doses o indi idual
analgesics o low doses o combina ions o analgesics
[3]. Combina ions o d ugs o high e icacy include
pa ace amol and codeine [4], pa ace amol and oxycodone
[5], ibup o en and codeine [6], ibup o en and oxycodone
[7], and ibup o en and pa ace amol [8]. E en adding
ca eine can imp o e analgesic e icacy as a combina ion
wi h con en ional analgesics [9].
The e icacy o combina ion analgesics has been shown
o be he sum o he e icacies o he indi idual analgesic
componen s, and was b oadly ue ac oss a ange o
di e en d ug combina ions, in pos ope a i e pain and
mig aine headache, and when es ed in he same and
di e en ials [10]. This means ha he e icacy o any
p oposed combina ion can be assessed heo e ically be o e
clinical ials a e conduc ed.
A po en ial pa o any combina ion migh be a as -
ac ing non-s e oidal an i-in lamma o y d ug (NSAID)
o mula ion, because speed o abso p ion and onse
p oduces good and long las ing analgesia [11, 12].
Dexke op o en is e ec i e in acu e pain a low doses
[13], and is also e ec i e in a wide a ie y o pain con-
di ions [14]; dexke op o en is he ac i e chi al o m
o ke op o en. T amadol is a widely used opioid o
p o en e icacy in combina ion wi h pa ace amol [15].
This s udy he e o e aimed a e alua ing he supe io
analgesic e icacy and sa e y o dexke op o en ome amol
and amadol hyd ochlo ide gi en as ou di e en ixed
combina ions and as single componen s in compa ison o
placebo, on mode a e o se e e acu e pain ollowing
impac ed hi d mandibula mola oo h ex ac ion. I was
also in ended o selec he op imum dose combina ion(s)
o be u he e alua ed in he subsequen phase III pi o al
s udies.
We used a o mula de i ed empi ically o es ima e
he possible e icacy ha migh be ob ained om dexke o-
p o en and amadol dosing combina ions [10]. The e
we e limi ed da a o dexke op o en om a sys ema ic
e iew [13] and wo indi idual pa ien le el analyses
o amadol [16, 17]. Es ima ed NNTs o combina ions
wi h dexke op o en ome amol 25 mg and amadol
hyd ochlo ide 37.5 mg o 75 mg we e 2.2 and 1.6,
espec i ely. Es ima es o combina ions wi h lowe
dexke op o en ome amol doses (12.5 mg) we e highe
(wo se) han 3 o abo e. The e was limi ed con idence
in he NNT es ima es o he combina ions due o
unce ain y in he e icacy es ima es o indi idual d ugs
because o low numbe s.
Me hods
The s udy (Sponso Code DEX-TRA-02; Eud aCT num-
be 2010-022798-32) was egis e ed a clinical ials.go
(NCT01307020). I was pe o med a 16 s udy si es in six
Eu opean coun ies (Ge many, Hunga y, I aly, Poland,
Spain and he Uni ed Kingdom). I was conduc ed in
acco dance wi h he p inciples o Good Clinical P ac ice
and he Decla a ion o Helsinki and was app o ed by
all he conce ned Compe en Au ho i ies and E hics
Commi ees. All pa icipa ing pa ien s p o ided w i en
in o med consen . The clinical phase o he s udy s a ed
on 23 d Feb ua y 2011 ( i s pa ien sc eened) and
concluded on 14 h Oc obe 2011 (las pa ien ou ).
Pa ien s
Heal hy male o emale pa ien s, aged 18 o 70 yea s, we e
eligible o he s udy i hey we e scheduled o ou pa ien
su gical emo al, unde local anaes hesia, o one o mo e
hi d mola s, a leas one o which was ully o pa ially
impac ed in mandibula bone. C i e ia o andomisa ion
included pos ope a i e pain o mode a e o se e e in ensi y
(Visual Analogue Scale [VAS] ≥40 mm and 4-poin Ve bal
Ra ing Scale [VRS] ≥2) wi hin ou hou s a e su ge y.
Pa ien s we e excluded om he s udy in any o he
ollowing ci cums ances: p egnan o b eas eeding women
o women o child-bea ing po en ial no using adequa e
con acep ion; known alle gy o he s udy d ugs, pa ace a-
mol, ace ylsalicylic acid, opioids o o he NSAIDs; mode a e
o se e e enal, hepa ic o ca diac dys unc ion; his o y o
gas oin es inal diso de s, bleeding diso de s; epilepsy,
as hma, angioedema o ela ed diso de s; his o y o d ug o
alcohol abuse; p esence o any medical condi ion ha in he
opinion o he in es iga o migh pose a isk o he pa ien ,
may con ound s udy esul s o migh impai compliance
wi h he s udy p ocedu es. Pa ien s who had ecei ed any
in es iga ional d ug o pa icipa ed in any o he clinical ial
wi hin he p e ious mon h we e also excluded. Fu he
exclusion c i e ia included signi ican su gical complica-
ions, o e all su ge y du a ion longe han one hou and
need o e-anaes hesia. Pa ien s who had aken any analge-
sics less han 24 h be o e su ge y we e also excluded.
Concomi an use o alcohol, psychoac i e d ugs, seda i es
and any o he medica ions o he apies ha could pose a
isk o he pa ien o con ound he s udy esul s we e no
pe mi ed wi hin 48 h and wo weeks be o e su ge y (de-
pending on he hal li e o he espec i e medica ions) and
up o 24 h pos -dose. Local applica ion o ice and he in ake
o ca eine we e no pe mi ed du ing he 24-hou pos -
dose pe iod. Pa ien s had o be in as ing condi ions om
wo hou s be o e su ge y and up o h ee hou s pos -dose.
S udy design
This was a mul icen e, andomised, double-blind, double-
dummy, pa allel-g oup, placebo-con olled, single-dose,
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 2 o 13
phase II, dose- inding s udy, wi h a o al o 10 ea men
a ms (wi h balanced alloca ion a io), including dex-
ke op o en ome amol (12.5 mg and 25 mg) and
amadol hyd ochlo ide (37.5 mg and 75 mg) gi en
as ou di e en ixed combina ions (DKP12.5/TRAM37.5;
DKP12.5/TRAM75; DKP25/TRAM37.5; DKP25/TRAM75)
and as single componen s (DKP12.5; DKP25; TRAM37.5;
TRAM75). An ac i e con ol (ibup o en 400 mg, as an acid
o mula ion) was included in o de o demons a e he
sensi i i y o he pain model, because i was signi ican ly
supe io o placebo in ials in he same indica ion [18–20],
and has he la ges body o da a o he indica ion [3].
The o e all s udy du a ion was app oxima ely 30 days
o each pa ien , including h ee isi s o he s udy si e:
Visi 1, o sc eening (wi hin wo weeks o hei scheduled
su ge y day); Visi 2, o den al su ge y, andomisa ion
and ea men adminis a ion, ollowed by a 24-hou
pos -dose pain and analgesic e ec assessmen pe iod
(wi h he i s ou hou s a he s udy si e), du ing
which pa ien s had o eco d e icacy da a using an
elec onic dia y (eDia y); and Visi 3 (End o S udy),
o inal sa e y ollow-up (10 ± 3 days a e su ge y day). In
addi ion, pa ien s ecei ed a phone call o sa e y assess-
men he day a e Visi 2 (app oxima ely 24 h pos -dose).
Fo hose pa ien s who me he selec ion c i e ia, he
su gical p ocedu e was pe o med unde s anda dised
local anaes hesia, which was limi ed o local anaes he ic
block using lidocaine (2 %) wi h epineph ine (1:80.000)
up o a o al olume o 5.4 mL pe mola . A e su ge y,
pa ien s epo ing pain we e asked o a e hei pain
in ensi y (PI) by a VAS (0–100; wi h he le end labelled
“no pain”and he igh end labelled “wo s possible
pain”; [21, 22]) and by a 4-poin VRS (0 = none, 1 = mild,
2 = mode a e, 3 = se e e; [21]) o assess hei eligibili y
o andomisa ion. Pa ien s expe iencing pain o mode a e
o se e e in ensi y (VAS ≥40 mm and VRS ≥2) wi hin ou
hou sa e heendo su ge ywe e andomisedand
ecei ed one single o al dose o he assigned s udy
ea men . A e andomisa ion and immedia ely p io
o he adminis a ion o s udy medica ion, VRS-PI was
measu ed again and he sco e was eco ded as baseline PI
o he e icacy analysis.
Pa icipan s we e andomly assigned o one o 10
ea men g oups ollowing a blocked andomisa ion
p ocedu e, wi h a block size o 10 and an alloca ion a io
o 1:1:1:1:1:1:1:1:1:1. The andomisa ion p ocess was
cen alised by an In e ac i e Voice/Web Response
Sys em (IVRS/IWRS) and he ea men code was
deli e ed o each pa ien acco ding o a compu e -
gene a ed andom alloca ed sequence ( andomisa ion lis )
p epa ed by a Sponso ’s hi d pa y p io o he s a o
he s udy. Two se s we e p epa ed, one se was used o
p og amming he IVRS/IWRS and he o he se was
used o he labelling o he s udy medica ion. Pe sonnel
in ol ed in he p epa a ion o he handling o he an-
domisa ion lis we e no in ol ed in he s udy conduc
and s a is ical analysis. Double-blind condi ions we e
secu ed by he iden ical appea ance and weigh o he
eigh es ed s udy d ugs as able s as well as he placebo
able ma ching he es ed s udy d ugs. In o de o keep
he ac i e con ol ibup o en blinded, he e was also a
placebo able ma ching ibup o en (each single-dose
ea men consis ed o wo able s) leading o a double-
dummy design. The blind was main ained o pa ien s and
o people esponsible o he ongoing conduc o he
s udy (such as he managemen , moni o s, in es iga o s)
and hose esponsible o da a analysis and in e p e -
a ion o esul s a he conclusion o he s udy, such
as biome ics pe sonnel.
Rescue medica ion (RM) consis ing o pa ace amol 1 g
(wi h a maximum ecommended daily dose o 4 g) was
a ailable on eques du ing he 24-hou pos -dose pe iod.
Pa ien s we e encou aged bu no compelled o wai
o a leas 60 min pos -dose, o allow ime o he
s udy medica ion o ake e ec .
E icacy e alua ion
Following ea men adminis a ion, pa ien s we e e-
ques ed o make mul iple assessmen s o pain in ensi y
and pain elie (PAR) on he eDia y o e a pe iod o 24 h.
They also had o make an o e all assessmen o he s udy
medica ion (pa ien global e alua ion, PGE) a he
end o his pe iod. The ime when RM was i s used,
i applicable, was also eco ded.
PI was measu ed on a 4-poin VRS (0 = none, 1 = mild,
2 = mode a e, 3 = se e e; [21]) immedia ely p io he
adminis a ion o s udy medica ion (baseline PI) and
hen a 15 min, 30 min, 45 min, 1 h, 1.5 h, 2 h, 2.5 h,
3 h, 3.5 h, 4 h, 5 h, 6 h, 8 h, 12 h and 24 h pos -dose. PAR
was measu ed on a 5-poin VRS (0 = none, 1 = sligh ,
2 = mode a e, 3 = good, 4 = comple e; [21]) a he
same p e-de ined pos -dose ime poin s. PGE was
measu ed on a 5-poin VRS (1 = poo , 2 = ai , 3 = good,
4 = e y good, 5 = excellen ; [21, 23]) a 24 h pos -dose (o
whene e he pa ien used RM, i his occu ed i s ).
When pa ien s used RM, a inal PI and PAR assessmen
and he PGE we e eco ded immedia ely be o e he in ake
and, a e ha , hey we e excluded om u he e icacy
measu emen s. A e use o RM, he baseline obse a ion
ca ied o wa d (BOCF) me hod was applied [24], wi h PI
e u ning o i s baseline sco e and PAR o ze o o all
subsequen ime poin s. I a pa ien p ema u ely wi hd ew
om he s udy, inal PI and PAR assessmen and he PGE
we e also eques ed.
F om he PI and PAR sco es, he summed pain in ensi y
di e ences (SPID) and he o al pain elie (TOTPAR)
o e 4, 6, 8 and 12 h pos -dose we e calcula ed. SPID was
calcula ed as he ime-weigh ed sum o he pain in ensi y
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 3 o 13
di e ence (PID) alues om baseline and TOTPAR was
calcula ed as he ime-weigh ed sum o he PAR sco es.
The pe cen ages o he heo e ical maximum possible
SPID (% max SPID) and o he heo e ical maximum
possible TOTPAR (% max TOTPAR) we e also calcula ed.
The p ima y e icacy endpoin was he pe cen age o
pa ien s showing esponse, de ined as he achie emen
o a leas 50 % o he maximum possible TOTPAR
(≥50 % max TOTPAR), o e 6 h pos -dose wi hin he
espec i e ea men a m [3, 24].
Seconda y e icacy endpoin s included: pe cen age o
esponde s (≥50 % max TOTPAR) o e 4, 8 and 12 h;
mean PI and PAR (VRS) sco es o e 24 h; SPID, % max
SPID, TOTPAR and % max TOTPAR o e 4, 6, 8 and 12 h;
PGE a he end o he assessmen pe iod; ime o i s use
o RM since ea men adminis a ion and pe cen age o
pa ien susingRMo e 4,6,8,12and24h.
Sa e y e alua ion
The sa e y e alua ion was based on he incidence,
se iousness, in ensi y and causal ela ionship o ea men -
eme gen ad e se e en s (AEs). AEs we e assessed
h oughou he en i e s udy by means o a non-leading
open ques ion. Spon aneously epo ed AEs we e also
eco ded. Fu he mo e, sa e y was also e alua ed by he
assessmen o clinically signi ican changes pos -dose
e sus baseline in physical examina ion, i al signs
(VS; blood p essu e and hea a e), 12-lead elec oca dio-
g am (ECG) and labo a o y sa e y es s (haema ology,
biochemis y and u inalysis). Any pa ien who p ema-
u ely wi hd ew a e ha ing ecei ed s udy medica ion
was encou aged o unde go Visi 3.
S a is ical analysis
Fo he p ima y e icacy a iable, in o de o demons a e
he supe io i y o ac i e ea men in compa ison wi h
placebo he null hypo hesis o equali y be ween placebo
and each es ed s udy d ug ( he ou combina ions and
he ou co esponding single agen s) was es ed using a
Chi-squa e es . Mul iplici y was adjus ed by using he
Šidák co ec ion [α=1-(1- α) ^ (1/k)], whe e k was he
numbe o compa isons. Conside ing eigh compa isons, a
ype I e o p obabili y = 0.00639 was used o he single
compa isons. The null hypo hesis o equali y be ween
placebo and he ac i e con ol was also es ed o alida e
he pain model. In addi ion, e en a es (ER), ela i e isk
(RR), NNT and ela i e isk educ ion (RRR), wi h
hei co esponding 95 % con idence in e als (CI),
we e es ima ed in o de o compa e he e ec size o
placebo wi h he e ec size o each ac i e ea men .
Seconda y e icacy a iables we e analysed as ollows:
pe cen age o esponde s o e 4, 8 and 12 h we e
analysed analogously o he p ima y e icacy a iable;
mean PI and PAR (VRS) sco es we e analysed by
means o desc ip i e s a is ics; quan i a i e a iables
(TOTPAR, % max TOTPAR, SPID and % max SPID)
showing homogenei y o a iance (acco ding o Le ene’s
es ) we e analysed by one-way analysis o a iance
(ANOVA) using he Dunne ’s es o compa ison be ween
placebo and each ac i e ea men , wi h an o e all signi i-
cance le el o 5 % wo-sided; o dinal a iables (PGE) and
quan i a i e a iables showing no homogenei y o a iance
we e analysed by he Wilcoxon ank-sum es o
compa ison be ween placebo and each ac i e ea men ,
using he Hochbe g co ec ion o he adjus men o
mul iple compa isons, wi h an o e all signi icance le el o
5 % wo-sided; ime o RM was analysed using he
Kaplan-Meie es ima ion me hod and ea men g oups
we e compa ed using a log- ank es , wi h he Hochbe g
me hod applied o he mul iplici y co ec ion; pe cen age
o pa ien s using RM we e analysed analogously o he
p ima y e icacy a iable. Sa e y a iables we e analysed by
means o desc ip i e s a is ics.
As an explo a o y analysis, he p ima y e icacy endpoin
was eassessed using ac i e con ol as compa a o . The
s a is ics es was used o e alua e i he e ec o dexke o-
p o en, amadol and hei combina ion was s a is ically
signi ican on he ou come. When a leas one o he
h ee es s was ound o be s a is ically signi ican ,
he Tukey me hod was applied o ind ou which
doses ga e a signi ican di e ence on he ou come.
S a is ical di e ences be ween NNTs we e examined
using he z- es [25].
All e icacy analyses we e pe o med on he “in en ion-
o- ea ”(ITT) popula ion ( andomised pa ien s who
ecei ed s udy medica ion and o whom a leas one
pos -dose assessmen was a ailable). The “pe p o ocol”
(PP) popula ion (pa ien s o he ITT popula ion wi h no
majo p o ocol iola ions) was used o pe o m con i ma-
o y analyses on he p ima y endpoin . Sa e y analyses
we e pe o med on he “sa e y”popula ion ( andomised
pa ien s who ecei ed s udy medica ion).
I was es ima ed ha 540 e aluable pa ien s (60 pe
ea men a m) would ha e o be included in o de o
achie e app oxima ely 80 % powe in ejec ing he null
hypo hesis o equali y be ween placebo and he eigh
expe imen al ea men a ms ( he ou combina ions
and he ou co esponding single agen s) ega ding
he p ima y endpoin , on he basis o he ollowing
assump ions [26]: esponse a e o placebo = 0.13; expec ed
RR o esponse in ac i e ea men e sus placebo = 3.21;
o e all ype I e o p obabili y o 0.05 ( wo-sided). Six y
u he pa ien s we e o be ea ed wi h he ac i e con ol
as a e e ence o he e ec o he eigh expe imen al
ea men a ms. I was expec ed ha app oxima ely 667
pa ien s would ha e o be sc eened in o de o ob ain 600
andomised pa ien s, assuming an app oxima e a e o
10 % sc eening ailu es.
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 4 o 13
Resul s
O he 745 pa ien s sc eened, 611 pa ien s we e an-
domised and ecei ed he s udy ea men , hus con-
s i u ing he sa e y popula ion. E icacy analyses we e
pe o med on he ITT popula ion o 606 andomised
pa ien s. The PP popula ion o 567 pa ien s was used
o pe o m con i ma o y analyses on he p ima y end-
poin . Pa ien assignmen o he di e en popula ions
occu ed be o e he s udy blind was b oken. The pa ici-
pan low wi h he numbe s o pa icipan s who we e
andomly assigned, ecei ed in ended ea men , and
we e analysed o he p ima y ou come is ep esen ed
in Addi ional ile 1.
Demog aphy and baseline cha ac e is ics o di e en
ea men g oups we e compa able. Demog aphic and
baseline cha ac e is ics o he ITT popula ion a e ep e-
sen ed in Table 1. The o e all mean age was 27 yea s
( ange 18–64 yea s), 59 % we e women, and 90 % we e
whi e. The mean su ge y ime was 29 min, mos pa ien s
(90 %) had only one mandibula mola emo ed, and he
o al numbe o mola ex ac ions was one o wo in
95 % o pa ien s. Ini ial pain (baseline PI) was mode a e
o se e e in 601 pa ien s (Table 2).
Fou (0.65 %) pa ien s, ou o 611 andomised, discon-
inued he s udy a e andomisa ion, hus esul ing in a
o al o 607 pa ien s comple ing he s udy. One pa ien
(alloca ed o DKP12.5/TRAM75) discon inued he
s udy due o “ ailu e o eDia y”. The o he h ee pa ien s
(alloca ed espec i ely o DKP25/TRAM37.5, DKP25/
TRAM75 and TRAM37.5) we e “los o ollow-up”. Th ee
o hese ou pa ien s a ended Visi 3.
E icacy esul s
P ima y endpoin
The pe cen age o pa ien s wi h ≥50 % max TOTPAR
o e six hou s pos -dose was signi ican ly supe io o pla-
cebo o all DKP/TRAM combina ions and also o
DKP25 (p< 0.0001 o each compa ison, excep p=0.0009
o DKP12.5/TRAM37.5), wi h he highes pe cen age
o esponde s achie ed in he DKP25/TRAM75 g oup
(72 % e sus 10 % in he placebo g oup; Fig. 1 and
Addi ional ile 2.
De ailed ER, RR and NNT esul s a six hou s a e
p esen ed in Addi ional ile 3. DKP25/TRAM75 also
had he highes RR [7.2 (95 % CI: 3.3 o 15.7)] and
he lowes NNT [1.6 (95 % CI: 1.3 o 2.1)]. No o he
poin es ima e o NNT was below 2.0 (Fig. 2). The
NNT o DKP25/TRAM75 was signi ican ly be e han
o he NNT alues (p< 0.05, z- es ), excep o DKP25/
TRAM37.5, DKP12.5/TRAM75 and DKP25.
The pe cen age o esponde s (≥50 % max TOTPAR)
was signi ican ly supe io o placebo o all DKP/TRAM
combina ions and o bo h doses o DKP in mono he apy
(p< 0.0001) o e ou hou s and i emained signi ican ly
supe io o DKP25/TRAM75, DKP12.5/TRAM75, DKP25/
TRAM37.5 (p< 0.0001) and DKP25 (p= 0.0012) o e eigh
hou s; and o DKP25/TRAM75 (p= 0.0002), DKP12.5/
TRAM75 (p= 0.0004) and DKP25/TRAM37.5 (p=0.0028)
o e 12 h. The highes pe cen age o esponde s o e
ou , eigh and 12 h was achie ed wi h DKP25/TRAM75
(79 %, 54 % and 38 % espec i ely) e sus 6.5 % wi h
placebo. Resul s a e ep esen ed in Addi ional ile 4
and Addi ional ile 5.
The ac i e con ol, ibup o en 400 mg, was s a is ically
supe io o placebo (p< 0.0001), hus alida ing he pain
model. Analyses un on he PP popula ion con i med
he p ima y e icacy esul s.
Seconda y endpoin s
The ime cou se o mean PAR and PI o e he whole
24 h pos -dose demons a ed apid onse o pain elie
wi h dexke op o en alone o in combina ion, and ha
he addi ion o amadol o dexke op o en esul ed bo h
in g ea e peak pain elie and g ea e pain elie o e
he longe e m, pa icula ly a imes longe han six
hou s pos dose (Fig. 3, and Addi ional ile 6).
The analysis o summa y e icacy measu es (SPID, % max
SPID, TOTPAR and % max TOTPAR; Addi ional ile 7,
Addi ional ile 8, Addi ional ile 9, Addi ional ile 10 and
Addi ional ile 11) showed ha all DKP/TRAM com-
bina ions and bo h doses o DKP in mono he apy
we e signi ican ly supe io o placebo (p< 0.01) excep o
DKP12.5 o e 12 h, wi h he bes esul s achie ed wi h
DKP25/TRAM75. The ime cou se o mean SPID and
mean TOTPAR a e ep esen ed in Addi ional ile 12 and
Addi ional ile 13.
The ime o RM was signi ican ly longe (p< 0.005) o
all ac i e ea men s (excep o bo h doses o TRAM.HCl
in mono he apy) han o placebo, wi h DKP12.5/TRAM75
and DKP25/TRAM75 p esen ing he longes alue
(median ime, [95 % CI]: 8.5 h [5.9 o 13.0] and 8.1 h
[6.3 o 13.4] espec i ely, e sus 1.4 h [1.2 o 1.8] in he
placebo g oup). Fig. 4 shows he p opo ion o pa ien s
emedica ing o e ime in each g oup, wi h addi ional
in o ma ion in Addi ional ile 14 and Addi ional ile 15.
The pe cen age o pa ien s equi ing RM (Addi ional
ile 16 and Addi ional ile 17) o e ou and six hou s was
signi ican ly smalle (p< 0.00639) o DKP12.5/TRAM75,
DKP25/TRAM37.5, and DKP25/TRAM75 (also o DKP25
o e ou hou s) han o placebo (o e six hou s 47 %,
40 %, and 38 % o each combina ion espec i ely e sus
73 % o placebo). The di e ence emained signi ican o
DKP25/TRAM75 o e eigh hou s (48 % e sus 73 %) and
o DKP25/TRAM37.5 o e 12 and 24 h (49 % e sus 74 %
o bo h ime poin s).
The PGE o he s udy medica ion a he end o he
assessmen pe iod is ep esen ed in Fig. 5. All ea -
men s (excep TRAM37.5) we e signi ican ly supe io
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 5 o 13

Table 1 Demog aphic and baseline cha ac e is ics (ITT popula ion)
DKP 12.5 mg +
TRAM 37.5 mg
DKP 12.5 mg +
TRAM 75 mg
DKP 25 mg +
TRAM 37.5 mg
DKP 25 mg +
TRAM 75 mg
DKP
12.5 mg
DKP 25 mg TRAM 37.5 mg TRAM 75 mg Ibup o en Placebo O e all
n 60 62 63 61 606059 59 6062606
Gende n (%) emale 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)
E hnic o igin n (%) Whi e 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)
O he 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 (yea s) 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)
ange 18–52 18–53 18–64 18–52 18–63 18–48 18–55 18–58 18–44 18–54 18–64
BMI (kg/m
2
) 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)
ange 18–30 18–35 18–32 18–30 18–30 15–30 18–30 18–34 18–31 18–29 18–35
Su ge y du a ion
c
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)
To al hi d mola
ex ac ions
d
n (%)
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)
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)
Lowe hi d mola
ex ac ions n (%)
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)
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;
b
n= 59;
c
ime is exp essed in minu es and seconds;
d
o al numbe o hi d mola ex ac ions, including also uppe hi d mola ee h
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 6 o 13
Table 2 PI be o e andomiza ion and be o e ea men adminis a ion (ITT popula ion)
DKP 12.5 mg +
TRAM 37.5 mg
DKP 12.5 mg +
TRAM 75 mg
DKP 25 mg +
TRAM 37.5 mg
DKP 25 mg +
TRAM 75 mg
DKP 12.5 mg DKP 25 mg TRAM 37.5 mg TRAM 75 mg Ibup o en Placebo O e all
n 60 62 63 61 60 6059 59 6062606
PI be o e andomiza ion
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 Mode a e 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)
Se e e 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 be o e ea men adminis a ion (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)
Mode a e 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)
Se e e 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 measu ed on a 4-poin VRS (0 = ‘none’ o 3 = ‘se e e’). Baseline PI e e s o he VRS-PI eco ded immedia ely p io o he adminis a ion o he s udy medica ion (in con as o VRS-PI measu ed be o e
andomisa ion o eligibili y pu poses). Baseline PI was mode a e o se e e in 601 pa ien s; i was epo ed as “mild”by 3 (0.5 %) pa ien s and esul s we e missing o 2 (0.3 %) pa ien s due o compila ion da a e o
on he eDia y
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 7 o 13
o placebo (p< 0.01), wi h he highes sco es in he
DKP25/TRAM75 g oup. The pe cen age o pa ien s wi h
‘good’ o ‘excellen ’PGE esponse was 79 % in he DKP25/
TRAM75 g oup e sus 11 % in he placebo g oup. The
pe cen age o pa ien s wi h ‘ e y good’and ‘excellen ’PGE
was 51 % in he DKP25/TRAM75 g oup e sus 4.8 % in
he placebo g oup.
The analysis o dose– esponse ela ionship be ween he
es ed s udy d ugs and he ac i e con ol showed ha
DKP25/TRAM75 was he only combina ion ha was
signi ican ly supe io o ibup o en 400 mg (p= 0.0028).
Sa e y esul s
O 611 pa ien s ea ed, 40 (6.5 %) epo ed a o al o
63 ad e se eac ions (ADRs), he mos equen being
omi ing (21 pa ien s; 3.4 %), nausea (14 pa ien s; 2.3 %),
dizziness (11 pa ien s; 1.8 %) and somnolence (5 pa ien s;
0.8 %) (Table 3). Apa om one case o “se e e”somno-
lence in he DKP12.5/TRAM75 g oup (1.6 %), all ADRs
we e conside ed “mild”(45 ADRs, 71 %) o “mode a e”
(17 ADRs, 27 %) in in ensi y.
Only one se ious ad e se e en (SAE) was epo ed in
one pa ien (alloca ed o he TRAM75 g oup), consis ing in
dizziness o mild in ensi y, which equi ed hospi aliza ion
o moni o ing and esol ed spon aneously. The e en was
assessed as “possibly ela ed” o he s udy medica ion.
Dizziness is a commonly epo ed ADR associa ed o
he use o amadol, occu ing in mo e han 10 % o
pa ien s, acco ding o he au ho ized Summa y o P oduc
Cha ac e is ics.
Fig. 2 NNT o ≥50 % max TOTPAR compa ed wi h placebo o e six hou s pos dose. Maximum TOTPAR co esponds o he heo e ical
maximum possible ime-weigh ed sum o he PAR sco es, measu ed on a 5-poin VRS (0 = ‘none’ o 4 = ‘comple e’). Ba s show 95 % con idence
in e al o NNT, wi h colou change as poin es ima e (No e ha TRAM37.5 was no signi ican ly be e han placebo)
Fig. 1 Pe cen age o pa ien s showing esponse (≥50 % max TOTPAR) o e 6 h pos -dose (P ima y Endpoin ). Maximum TOTPAR co esponds o
he heo e ical maximum possible ime-weigh ed sum o he PAR sco es, measu ed on a 5-poin VRS (0 = ‘none’ o 4 = ‘comple e’)
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 8 o 13
The highes incidence o ADRs was epo ed in he
TRAM75 g oup (17 %). This incidence was highe
han epo ed in g oups ecei ing amadol 75 mg in
combina ion (i.e. DKP12.5/TRAM75 and DKP25/
TRAM75) o which he incidence was o 9.5 % and
12 % o pa ien s, espec i ely.
No dea hs o o he signi ican AEs occu ed. No
pa ien discon inued because o AEs. The e we e no
clinically ele an changes in he VS, physical examina ion,
12-lead ECG o labo a o y sa e y es s e sus baseline.
O e all, all ea men s we e sa e and well ole a ed, wi h
all DKP/TRAM combina ions p esen ing a sa e y and
Fig. 4 Cumula i e equency (Kaplan-Meie es ima ion) o RM in ake (0–24 h)
Fig. 3 Time cou se o mean PAR sco es (0–24 h). PAR measu ed on a 5-poin VRS (0 = ‘none’ o 4 = ‘comple e’)
Moo e e al. The Jou nal o Headache and Pain (2015) 16:60 Page 9 o 13