Efficacy of diclofenac sodium softgel 100 mg with or without caffeine 100 mg in migraine without aura: a randomized, double-blind, crossover study


Peroutka, S.J.; Lyon, J.A.; Swarbrick, J.; Lipton, R.B.; Kolodner, K.; Goldstein, J.

Headache 44(2): 136-141

2004


Objective: A phase II, randomized, double-blind, crossover study was designed to evaluate the efficacy of 100-mg diclofenac sodium softgel (formulated using ProSorb technology) with or without 100-mg caffeine versus placebo in migraineurs during migraine attacks. Background: Diclofenac has been demonstrated to be an effective migraine treatment in several placebo-controlled studies. A rapidly absorbed softgel of diclofenac has been shown to be effective in the rapid relief of acute pain, and may have advantages in migraine treatment. In addition, caffeine has consistently been shown to increase both the efficacy and speed of onset of concurrently administered analgesics. The ability of caffeine to both enhance and accelerate analgesic effects has been documented with a variety of different medications (ie, aspirin, acetaminophen, ibuprofen, and ergotamine). Methods: The 3-period crossover study was designed to compare diclofenac softgel 100 mg, diclofenac softgel 100 mg plus caffeine 100 mg, and placebo in the acute treatment of migraine. Subjects treated one moderate or severe attack with each study medication. The primary efficacy parameter was the percentage of subjects with headache relief at 60 minutes as defined by a reduction of headache severity from moderate or severe at baseline to absent or mild compared with placebo. Though the sample size estimate required that 72 subjects treat 3 separate attacks, 51 subjects treated 1 migraine attack, 44 treated 2 attacks, and 39 treated 3 attacks. Results: In the placebo group, 6 (14%) of 43 subjects reported headache relief at 60 minutes versus 12 (27%) of 45 subjects in the diclofenac softgel group, and 19 (41%) of 46 subjects in the diclofenac softgel plus caffeine group. Differences were statistically significant for the diclofenac softgel plus caffeine group versus placebo (odds ratio, 4.2; 95% confidence interval, 1.3 to 13.7). Rescue medication was used by 27 (63%) of 43 subjects treated with placebo, 15 (33%) of 45 subjects treated with diclofenac softgel, and 14 (30%) of 46 subjects treated with diclofenac softgel plus caffeine. This result is highly statistically significant (chi22=11.56, P=.003). Both the diclofenac plus caffeine (P<.03) and diclofenac only (P<.03) groups were significantly different from the placebo group in terms of the visual analog scale score at 60 minutes. Conclusions: The major finding of the present study is that diclofenac softgel plus caffeine produces statistically significant benefits relative to placebo at 60 minutes. Diclofenac softgel alone did not differ significantly from placebo, perhaps due to limits in sample size. Nonsignificant trends support the analgesic adjuvant benefit of caffeine when added to diclofenac softgels.

Efficacy
of
Diclofenac
Sodium
Softgel
100
mg
With
or
Without
Caffeine
100
mg
in
Migraine
Without
Aura:
A
Randomized,
Double-blind,
Crossover
Study
Stephen
J.
Peroutka,
MD,
PhD;
James
A.
Lyon,
PharmD;
James
Swarbrick,
DSc,
PhD;
Richard
B.
Lipton,
MD;
Ken
Kolodner,
ScD;
Jerome
Goldstein,
MD
Objective.—A
phase
II,
randomized,
double-blind,
crossover
study
was
designed
to
evaluate
the
efficacy
of
100-mg
diclofenac
sodium
softgel
(formulated
using
ProSorb
technology)
with
or
without
100-mg
caffeine
versus
placebo
in
migraineurs
during
migraine
attacks.
Background.—Diclofenac
has
been
demonstrated
to
be
an
effective
migraine
treatment
in
several
placebo-
controlled
studies.
A
rapidly
absorbed
softgel
of
diclofenac
has
been
shown
to
be
effective
in
the
rapid
relief
of
acute
pain,
and
may
have
advantages
in
migraine
treatment.
In
addition,
caffeine
has
consistently
been
shown
to
increase
both
the
efficacy
and
speed
of
onset
of
concurrently
administered
analgesics.
The
ability
of
caffeine
to
both
enhance
and
accelerate
analgesic
effects
has
been
documented
with
a
variety
of
different
medications
(ie,
aspirin,
acetaminophen,
ibuprofen,
and
ergotamine).
Methods.—The
3-period
crossover
study
was
designed
to
compare
diclofenac
softge1100
mg,
diclofenac
softgel
100
mg
plus
caffeine
100
mg,
and
placebo
in
the
acute
treatment
of
migraine.
Subjects
treated
one
moderate
or
severe
attack
with
each
study
medication.
The
primary
efficacy
parameter
was
the
percentage
of
subjects
with
headache
relief
at
60
minutes
as
defined
by
a
reduction
of
headache
severity
from
moderate
or
severe
at
baseline
to
absent
or
mild
compared
with
placebo.
Though
the
sample
size
estimate
required
that
72
subjects
treat
3
separate
attacks,
51
subjects
treated
1
migraine
attack,
44
treated
2
attacks,
and
39
treated
3
attacks.
Results.—In
the
placebo
group,
6
(14%)
of
43
subjects
reported
headache
relief
at
60
minutes
versus
12
(27%)
of
45
subjects
in
the
diclofenac
softgel
group,
and
19
(41%)
of
46
subjects
in
the
diclofenac
softgel
plus
caffeine
group.
Differences
were
statistically
significant
for
the
diclofenac
softgel
plus
caffeine
group
versus
placebo
(odds
ratio,
4.2;
95%
confidence
interval,
1.3
to
13.7).
Rescue
medication
was
used
by
27
(63%)
of
43
subjects
treated
with
placebo,
15
(33%)
of
45
subjects
treated
with
diclofenac
softgel,
and
14
(30%)
of
46
subjects
treated
with
diclofenac
softgel
plus
caffeine.
This
result
is
highly
statistically
significant
(x
=
11.56,
P
=
.003).
Both
the
diclofenac
plus
caffeine
(P
<
.03)
and
diclofenac
only
(P
<
.03)
groups
were
significantly
different
from
the
placebo
group
in
terms
of
the
visual
analog scale
score
at
60
minutes.
Conclusions.—The
major
finding
of
the
present
study
is
that
diclofenac
softgel
plus
caffeine
produces
statis-
tically
significant
benefits
relative
to
placebo
at
60
minutes.
Diclofenac
softgel
alone
did
not
differ
significantly
from
placebo,
perhaps
due
to
limits
in
sample
size.
Nonsignificant
trends
support
the
analgesic
adjuvant
benefit
of
caffeine
when
added
to
diclofenac
softgels.
Key
words:
migraine,
diclofenac,
ProSorb
technology,
caffeine
Abbreviations:
VAS
visual
analog
scale,
AEs
adverse
events
(Headache
2004;44:136-141)
From
Burlingame,
Calif
(Dr.
Peroutka);
aaiPharma
Inc,
Wilmington,
NC
(Drs.
Lyon
and
Swarbrick);
the
Departments
of
Neurology,
Epidemiology,
and
Social
Medicine,
Albert
Einstein
College
of
Medicine,
Bronx,
NY
and
Innovative
Medical
Research,
Stamford,
Conn
(Dr.
Lipton);
Innovative
Medical
Research,
Baltimore,
Md
(Dr.
IColodner);
and
San
Francisco
(Calif)
Clinical
Research
Headache
(Dr.
Goldstein).
Address
all
correspondence
to
Dr.
Stephen
J.
Peroutka,
1025
Tournament
Drive,
Burlingame,
CA
94010.
Accepted
for
publication
October
22,
2003.
136
Headache
137
Diclofenac
is
a
nonsteroidal
anti-inflammatory
drug
(NSAID)
with
potent
anti-inflammatory,
anal-
gesic,
and
antipyretic
properties.
Its
mechanism
of
ac-
tion
is
associated
principally
with
the
inhibition
of
prostaglandin
synthesis
(specifically,
inhibition
of
cy-
clooxygenase).
1
Diclofenac
is
indicated
for
a
vari-
ety
of
conditions
such
as
acute
and
chronic
arthri-
tis,
rheumatoid
arthritis,
osteoarthritis,
and
ankylosing
spondylitis.
Diclofenac
also
has
been
demonstrated
to
be
an
effective
migraine
treatment
in
several
placebo-
controlled
studies.
2-7
Migraine
is
defined
as
a
disorder
associated
with
intermittent
attacks
of
headache
(which
last
from
4
to
72
hours),
accompanied
by
autonomic
and
neu-
rologic
dysfunction.
8
In
some
subjects
and
in
some
attacks,
aura
precedes
the
headache.
Aura
occurs
as
neurologic
deficiency
or
irritation
symptoms
like
scintillating
scotoma,
hemianopsia,
paresthesia,
mo-
tor
weakness,
and
dysphasia.
At
least
a
component
of
migraine
is
presumed
to
involve
the
release
of
neuropeptides
and
inflammatory
mediators,
possibly
leading
to
a
neurogenic
inflammation
of
the
dura
vessels.
Caffeine
is
used
therapeutically
to
treat
somno-
lence,
postprandial
hypotension,
obesity,
and
respira-
tory
depression
in
neonates.
Caffeine
alone
may
have
some
mild
analgesic
effects.
9
It
is
a
common
ingredient
in
analgesic
preparations.
Indeed,
caffeine
has
been
found
to
be
an
effective
analgesic
adjuvant
when
used
in
combination
with
more
traditional
analgesics.
9-14
It
consistently
has
been
shown
to
increase
both
the
effi-
cacy,
as
well
as
the
speed
of
onset,
of
concurrently
ad-
ministered
analgesics.9,
10,13,14
This
ability
of
caffeine
to
both
enhance
and
accelerate
analgesic
effects
has
been
documented
with
a
variety
of
different
medica-
tions
(ie,
aspirin,
acetaminophen,
ibuprofen,
and
er-
gotamine).
7-12
In
addition,
caffeine
combination
prod-
ucts
have
been
shown
to
be
effective
in
the
acute
treat-
ment
of
migraine.
15,16
Relatively
few
studies,
however,
have
investigated
the
effect
of
caffeine
alone
in
the
treatment
of
pain.
In
a
study
of
analgesic
efficacy
in
tension-type
headache,
65-mg
caffeine
was
found
to
have
analgesic
effects
that
were
equivalent
to
648-
mg
acetaminophen.
17
The
analgesic
effect
of
caffeine
was
independent
of
its
effects
on
mood
or
on
caffeine
withdrawal.
Other
studies
have
reported
no
significant
therapeutic
effect
of
caffeine
in
pain
conditions
such
as
migraine.
14
Therefore,
the
present
study
was
designed
to
de-
termine
the
antimigraine
efficacy
of
100-mg
diclofenac
sodium
softgel
with
or
without
100-mg
caffeine
versus
placebo
in
migraineurs
during
migraine
attacks.
METHODS
This was
a
randomized,
double-blind,
3-period,
crossover
study
to
evaluate
the
efficacy
of
100-mg
di-
clofenac
and
that
of
100-mg
diclofenac
plus
100-mg
caffeine
versus
placebo
in
migraineurs
during
an
at-
tack
without
aura.
Subjects
18
to
60
years
of
age
with
a
diagnosis
of
migraine,
according
to
International
Headache
Society
criteria,
for
at
least
12
months
be-
fore
the
study
were
identified.
Subjects
with
aura
also
had
to
experience
attacks
without
aura.
Seventy-two
subjects
were
enrolled
in
the
single-center
study
during
a
4-month
recruitment
period.
The
trial
was
conducted
under
ambulatory
conditions.
Written
informed
con-
sent
was
obtained
from
all
subjects.
The
study
was
con-
ducted
under
Good
Clinical
Practice
and
International
Conference
on
Harmonization
of
Technical
Require-
ments
for
Registration
of
Pharmaceuticals
for
Human
Use
guidelines.
Clinical
End
Points.—The
primary
end
point
was
defined
as
a
1-hour
headache
response
(ie,
a
pain
re-
duction
to
mild
or
none
from
a
baseline
of
moderate
or
severe).
This
response
was
chosen
as
the
primary
end
point
to
determine
if
diclofenac,
alone
or
in
combi-
nation,
could
provide
rapid
and
significant
relief.
The
secondary
clinical
end
points
included
a
visual
ana-
log
scale
(VAS)
severity
rating
at
60
minutes
using
all
headaches,
frequency
of
headache
recurrence,
and
res-
cue
medications
used
within
6
hours
posttreatment.
Statistical
Analysis.—A
sample
size
of
72
subjects
(12
in
each
of
6
treatment
sequences)
was
selected
because
it
was
estimated
to
be
sufficient
to
have
80%
power
to
detect
a
20%
difference
in
proportion
of
clin-
ical
success
in
subjects.
The
calculation
is
based
on
the
McNemar
test
under
assumption
that
the
proportion
of
the
discordant
pairs
equal
38%.
To
evaluate
the
success
of
the
randomization,
cross-tabulations
with
chi-square
analysis
were
used
to
compare
treatment
groups
at
each
time
period.
To
compare
treatment
groups
for
the
primary
outcome,
138
February
2004
chi-square
and
logistic
regression
(presented
with
odds
ratios
[OR]
and
95%
confidence
intervals
[CI])
was
used
for
each
treatment
period.
To
compare
treatment
groups
for
all
treated
headaches
(combining
all
pe-
riods),
several
modeling
strategies
were
used.
First,
we
investigated
whether
there
was
a
carry-over
effect.
There
was
no
evidence
of
a
carry-over
effect,
as
ex-
pected,
since
the
plasma
half-life
of
both
diclofenac
and
caffeine
is
only
approximately
2
hours.
Accordingly,
we
did
not
adjust
for
carry
over
in
the
final
models.
General
estimated
equations
(GEEs)
were
used
on
all
available
headaches
to
control
for
within-subject
correlation.
18
The
analysis
on
those
sub-
jects
who
had
3
headache
diaries
(and
therefore
com-
pleted
the
study
according
to
protocol)
was
repeated.
Lastly,
we
repeated
the
analysis
using
a
different
mod-
eling
approach,
log-linear
models.
Baseline
severity
was
used
as
a
covariate
for
adjusted
models.
For
the
continuous
outcome
of
the
VAS,
headaches
over
all
periods
were
combined.
Analysis
of
covariance
was
used
to
compare
treatment
groups
where
the
baseline
VAS
score
and
a
subject
term
were
included
as
covariates.
The
subject
term
was
en-
tered
to
control
for
possible
within-subject
correlation.
All
analyses
were
performed
using
Statistical
Analysis
Systems
(SAS)
software,
version
6.12
(SAS
Institute,
Inc,
Cary,
NC,
1998).
RESULTS
The
study
was
powered
so
that
72
subjects
were
to
complete
3
headache
diaries.
Only
52
subjects,
how-
ever,
completed
the
first
diary,
46
completed
the
sec-
ond,
and
39
completed
the
third
diary.
One
subject
was
included
as
completing
the
second
and
third
diaries,
but
did
not
complete
the
first
diary.
Hence,
there
were
51
evaluable
first
headache
diaries.
Two
subjects
rated
their
second
headache
at
baseline
as
"mild."
Since
this
did
not
meet
protocol
specifications,
those
2
di-
aries
were
not
included
in
the
analysis,
resulting
in
44
subjects
who
completed
the
second
diary
according
to
the
protocol.
The
end
result
was
that
36
subjects
com-
pleted
all
3
diaries.
Subject
and
Headache
Characteristics.—Demo-
graphics
of
the
72
subjects
entered
into
the
study
are
presented
in
Table
1.
Average
age
of
the
participants
Table
1.—Subject
and
Headache
Characteristics*
Feature
Study
Group
(N
=
72)
Sex
Male
10
(13.9)
Female
62
(86.1)
Age,
y
1-39
20
(27.8)
4-49
27
(37.5)
50+
25
(34.7)
Race
White
56
(77.8)
Black
12
(16.7)
Asian
1
(1.4)
Other
3
(4.2)
Unilateral
pain
Yes
59
(81.9)
No
13
(18.1)
Pulsating
pain
Yes
53
(73.6)
No
18
(25.0)
Not stated
1
(1.4)
Moderate
to
severe
intensity
Yes
72
(100)
No
0
Worsened
by
physical
activity
Yes
62
(86.1)
No
10
(13.9)
Nausea/vomiting
Yes
62
(86.1)
No
10
(13.9)
Photophobia/phonophobia
Yes
71
(98.6)
No
1
(1.4)
*Values
are
numbers
(percentages).
was
44.7
years
(SD,
10.4)
with
an
age
range
of
24
to
64
years.
Table
2
shows
the
distribution
of
various
charac-
teristics
of
each
headache.
A
statistical
comparison
by
headache
was
not
made
(due
to
limited
power),
but
it
is
interesting
to
note
that
those
completing
a
third
diary
were
more
likely
to
report
severe
headache
intensity.
Primary
Efficacy
Measure:
Headache
Response
at
60
Minutes
For
All
Headaches.—Given
the
small
sample
and
the
dropout
rate
(diaries
2
and
3),
we
examined
all
headaches
together.
Ignoring
for
de-
scriptive
purposes
the
issue
of
correlated
data
(the
same
subjects
are
observed
over
more
than
one
headache),
Table
3
presents
a
descriptive
comparison
of
headache
response
data
for
the
treatment
groups
for
all
headache
diaries.
The
diclofenac
only
group
Headache
139
Table
2.-Characteristics
of
Headaches
Treated
During
3
Separate
Attacks*
Variable
First
Headache
(n
=
51)
Second
Headache
(n
=
44)
Third
Headache
(n
=
39)
Intensity
Moderate
40
(78.4)
32
(78.4)
20
(513)
Severe
11
(21.6)
12
(21.6)
19
(48.7)
Disability
Worked/functioned
normally
5
(9.8)
3
(6.8)
1
(2.6)
Mildly
impaired
29
(56.9)
21
(47.7)
17
(43.6)
Severely
impaired
12
(23.5)
15
(34.1)
12
(30.8)
Required
bed
rest
5
(9.8)
5
(11.4)
9
(23.1)
Nausea
Yes
17
(33.3)
25
(56.8)
23
(59.0)
No
34
(66.7)
19
(43.2)
16
(41.0)
Phonophobia
Yes
34
(66.7)
34
(77.3)
27
(69.2)
No
17
(33.3)
10
(22.7)
12
(30.8)
Photophobia
Yes
43
(84.3)
38
(86.4)
35
(89.7)
No
8
(15.7)
6
(13.6)
4
(103)
Vomiting
Yes
0
(0.0)
1
(2.3)
4
(103)
No
51
(100)
43
(97.7)
35
(89.7)
*Values
are
numbers
(percentages).
demonstrated
nearly
twice
the
response
when
com-
pared
with
the
placebo
group.
The
diclofenac
plus
caf-
feine
group
showed
nearly
triple
the
response.
Table
4
shows
the
results
of
the
GEEs
analysis
using
all
available
diaries.
The
results
were
statisti-
cally
significant
for
the
diclofenac
plus
caffeine
group,
but not
quite
significant
for
the
diclofenac
only
group.
Adjusting
for
the
baseline
severity
of
each
headache
did
not
affect
the
results
significantly.
An
analysis
us-
Table
3.-Comparison
of
Headache
Response
at
60
Minutes
for
Placebo,
Diclofenac
Plus
Caffeine,
and
Diclofenac
Only*
Yes
No
Total
Treatment
(n
=
37)
(n
=
97)
(n
=134)
Placebo
6
(14.0)
37
(86.0)
43
(32.1)
Diclofenac
plus
caffeine
19
(41.3)
27
(58.7)
46
(34.3)
Diclofenac
only
12
(26.7)
33
(73.3)
45
(33.6)
*Values
are
numbers
(percentages).
Table
4.-Results
of
General
Estimated
Equations
Analysis
of
Headache
Response
at
60
Minutes
for
Placebo,
Diclofenac
Plus
Caffeine,
and
Diclofenac
Only,
Unadjusted
and
Adjusted
for
Baseline
Severity
Treatment/Baseline
Severity
Odds
Ratio
(95%
Confidence
Interval)
Crude
Adjusted
Placebo
1.00
(-)
1.00
(-)
Diclofenac
plus
caffeine
4.28
(1.41-12.99)
4.22
(1.30-13.72)
Diclofenac
only
2.21
(.84-5.79)
2.48
(0.91-6.74)
Baseline
seventy
Severe
1.00
(-)
Moderate
3.86
(1.38-10.82)
ing
log-linear
models
produced
nearly
identical
results
(data
not
presented
here).
In
an
additional
analysis
us-
ing
GEE,
we
considered
only
those
subjects
who
com-
pleted
all
3
diaries
and
treatments
(36
subjects
con-
tributing
108
headaches).
The
effects
mirrored
previ-
ous
results,
but
the
ORs
were
somewhat
attenuated
and
became
nonsignificant
largely
due
to
the
reduc-
tion
in
power.
There
was
no
evidence
of
a
carry-over
effect,
as
would
be
expected,
since
the
plasma
half-life
of
both
diclofenac
and
caffeine
is
only
approximately
2
hours.
Secondary
Efficacy
Measures.-Visual
Analog
Scale
Severity
Rating
at
60
Minutes
For
All
Head-
aches-Table
5
shows
that
both
the
diclofenac
plus
caf-
feine
(P
<
.003)
and
diclofenac
only
(P
<
.03)
groups
were
significantly
different
from
the
placebo
group
in
terms
of
the
VAS
score
at
60
minutes.
The
diclofenac
plus
caffeine
group
shows
the
largest
effect.
Headache
Recurrence
For
All
Headaches.-
Headache
recurrence
is
defined
for
those
subjects
ex-
periencing
headache
relief
at
60
minutes
and
who
then
experienced
a
worsening
of
their
headache
to
moder-
ate
or
severe
within
24
hours.
Of
the
37
headaches
demonstrating
relief
at
60
minutes,
only
3
returned
to
moderate
or
severe
pain.
All
3
were
treated
with
res-
cue
medications.
Therefore,
no
subjects
qualified
as
experiencing
headache
recurrence.
Rescue
Medication
Within
First
6
Hours
Posttreat-
ment.-Of
134
treated
headaches,
56
(42%)
were
treated
with
rescue
medications
within
the
6-hour
140
February
2004
Table
5.—Results
of
Visual
Analog
Scale
(VAS)
Severity
Rating
at
60
Minutes
For
All
Headaches
(n
=130)
Treated
With
Placebo,
Diclofenac
Plus
Caffeine,
and
Diclofenac
Only*
P
Value
From
95%
Comparison
of
No.
of
Adjusted
Confidence
Treatment
Group
Treatment
Headaches
Meant
Interval
to
Placebo
Placebo
42
65.1
58.4-71.7
Diclofenac
+
caffeine
45
50.6
44.2-57.0
.0026
Diclofenac
only
43
54.9
483-61.4
.0315
*Four
diaries
were
missing
the
VAS
score
at
one
of
the
headaches
resulting
in
a
reduced
number
of
130
from
134
possible
headaches.
tBaseline
VAS
score
was
entered
as
a
covariate
along
with
a
"subject"
term.
The
overall
effect
of
treatment
was
F2,129
=
5.11,
P
=
.0083.
period.
Of
the
43
placebo-treated
headaches,
rescue
medications
were
used
for
27
headaches
(63%).
This
compares
to
30%
(14
of
46)
for
the
diclofenac
plus
caffeine
group
and
33%
(15
of
45)
for
the
diclofenac
only
group.
This
result
was
highly
statistically
signifi-
cant
(x
=
11.56,
P
=
.003).
Frequency
of
Adverse
Events.—Adverse
events
(AEs)
were
reported
in
10%
(14
of
134)
of
all
headaches.
Of
these,
one
AE
was
"possibly"
and
one
"probably"
related
to
the
drug.
For
the
43
placebo-
treated
headaches,
2
AEs
were
reported,
one
unre-
lated
and
one
possibly
related
to
treatment.
For
the
45
headaches
treated
with
diclofenac
only,
one
AE
was
reported
as
unrelated
to
the
treatment.
For
the
46
headaches
treated
with
diclofenac
plus
caffeine,
10
AEs
were
reported
as
unrelated
to
the
treatment,
and
one
as
probably
related.
The
apparent
higher
occur-
rence
of
total
AEs
for
those
headaches
treated
with
diclofenac
plus
caffeine
was
statistically
significant
(x
=
13.71,
P
=
.0011).
Nausea
was
reported
by
2
subjects.
No
serious
AEs
were
reported.
DISCUSSION
The
major
finding
of
the
present
study
is
that
di-
clofenac
softgel
plus
caffeine
produces
statistically
sig-
nificant
benefits
relative
to
placebo
at
60
minutes.
Di-
clofenac
softgel
alone
did
not
differ
statistically
from
placebo,
perhaps
due
to
limits
in
sample
size.
Given
the
magnitude
of
the
treatment
differences
observed,
however,
had
we
met
our
original
recruiting
and
treat-
ment
goals,
we
believe
that
the
results
would
have
been
statistically
significant.
The
nonsignificant
trends
sup-
port
the
analgesic
adjuvant
benefit
of
caffeine
when
added
to
diclofenac
softgels.
Overall,
the
results
of
this
study
confirm
earlier
observations
that
diclofenac
alone
is
more
effective
than
placebo
in
the
acute
treatment
of
migraine.
2-7
In
addition,
a
rapid
and
effective
analgesic
effect
of
diclofenac
sodium
softgel
has
been
reported
in
the
treatment
of
postoperative
dental
pain.
19
The
cur-
rent
results
further
suggest
that
the
combination
of
diclofenac
sodium
softgel
and
caffeine
is
more
effec-
tive
than
diclofenac
alone
in
the
acute
treatment
of
migraine.
Additional
studies
are
planned
to
further
evaluate
the
effectiveness
of
diclofenac
softgels
alone
since
the
present
study
was
limited
by
the
small
sample
size
and
high
dropout
rate.
Further
studies
would
use
a
more
conventional
standard
parallel
design
rather
than
a
crossover
design
to
aid
in
patient
enrollment.
In
summary,
the
current
results
show
that
headache
response
rates
at
60
minutes
were
14%
for
placebo,
27%
for
diclofenac
only,
and
41%
for
di-
clofenac
plus
caffeine.
Diclofenac
plus
caffeine
was
significantly
superior
to
placebo
in
1-hour
headache
response
(OR,
4.22;
95%
CI,
1.30
to
13.72).
Correcting
for
within-subject
correlation,
differences
in
headache
response
at
1
hour
between
diclofenac
and
placebo
did
not
quite
reach
statistical
significance
(OR,
2.48;
95%
CI,
0.91
to
6.74).
With
a
sample
size
as
originally
intended
and
assuming
a
similar
effect,
however,
the
diclofenac
only
group
would
have
produced
a
statisti-
cally
significant
result.
Headache
141
Analysis
of
the
VAS
severity
rating
supported
the
findings
above.
Both
diclofenac
groups
demonstrated
a
significant
effect
compared
to
the
placebo
group
when
all
headaches
were
considered.
The
size
of
the
effect
was
not
large
enough
given
the
sample
size
to
observe
a
significant
effect
when
we
viewed
the
data
for
an
individual
headache
(ie,
within
a
period).
These
results
suggest
that
diclofenac
softgels
and
diclofenac
softgels
plus
caffeine
may
represent
a
highly
efficacious
and
safe
therapeutic
approach
to
the
acute
treatment
of
migraine.
Further
studies
are
planned
to
assess
more
fully
the
clinical
efficacy
of
diclofenac
soft-
gels
in
the
acute
treatment
of
migraine.
REFERENCES
1.
Mitchell
JA,
Warner
TD.
Cyclo-oxygenase-2:
phar-
macology,
physiology,
biochemistry
and
relevance
to
NSAID
therapy.
Br
J
Pharmacol.
1999;128:1121-
1132.
2.
Del
Bene
E,
Poggioni
M,
Garagiola
U,
Maresca
V.
Intramuscular
treatment
of
migraine
attacks
using
di-
clofenac
sodium:
a
crossover
clinical
trial.
J
Int
Met
Res.
1987;15:44-48.
3.
Massiou
H,
Serrurier
D,
Lasserre
0,
Bousser
MG.
Effectiveness
of
oral
diclofenac
in
the
acute
treatment
of
common
migraine
attacks:
a
double-
blind
study
versus
placebo.
Cephalalgia.
1991;11:59-
63.
4.
Karachalios
GN,
Fotiadou
A,
Chrisikos
N,
Karabet-
sos
A,
Kehagioglou
K.
Treatment
of
acute
migraine
attack
with
diclofenac
sodium:
a
double-blind
study.
Headache.
1992;32:98-100.
5.
Dahlof
C,
Bjorkman
R.
Diclofenac-K
(50
and
100
mg)
and
placebo
in
the
acute
treatment
of
migraine.
Cephalalgia.
1993;13:117-123.
6.
The
Diclofenac-L/Sumatriptan
Migraine
Study
Group.
Acute
treatment
of
migraine
attacks:
efficacy
and
safety
of
a
nonsteroidal
anti-inflammatory
drug,
diclofenac-potassium,
in
comparison
to
oral
sumatriptan
and
placebo.
Cephalalgia.
1999;19:232-
240.
7.
McNeely
W,
Goa
KL.
Diclofenac-potassium
in
mi-
graine.
Drugs.
1999;6:991-1003.
8.
Headache
Classification
Committee
of
the
Interna-
tional
Headache
Society.
Classification
and
diagnos-
tic
criteria
for
headache
disorders,
cranial
neuralgias
and
facial
pain.
Cephalalgia.
1988;8(suppl
7):1-96.
9.
Laska
EM,
Sunshine
A,
Mueller
F,
Elvers
WB,
Siegel
C,
Rubin
A.
Caffeine
as
an
analgesic
adjuvant.
JAMA.
1984;251:1711-1718.
10.
Schmidt
R,
Fanchamps
A.
Effect
of
caffeine
on
in-
testinal
absorption
of
ergotamine
in
man.
Eur
J
Clin
Pharmacol.
1974;7:213-216.
11.
Jain
AK,
McMahon
FG,
Ryan
JR,
Unger
D,
Richard
W.
Aspirin
and
aspirin-caffeine
in
postpartum
pain
relief.
Clin
Pharmacol
Ther.
1978;24:69-75.
12.
Schachtel
BP,
Fillingim
JM,
Lane
AC,
Thoden
WR,
Baybutt
RI.
Caffeine
as
an
analgesic
adjuvant.
A
double-blind
study
comparing
aspirin
with
caffeine
to
aspirin
and
placebo
in
patients
with
sore
throat.
Arch
Intern
Med.
1991;151:733-737.
13.
Forbes
JA,
Beaver
WT,
Jones
KF,
et
al.
Effect
of
caffeine
on
ibuprofen
analgesia
in
postoperative
oral
surgery
pain.
Clin
Pharmacol
Ther.
1991;49:674-684.
14.
Sawynok
J.
Pharmacological
rationale
for
the
clinical
use
of
caffeine.
Drugs.
1995;49:37-50.
15.
Lipton
RB,
Stewart
WF,
Ryan
RE
Jr,
Saper
J,
Silberstein
S,
Sheftell
F.
Efficacy
and
safety
of
acetaminophen,
aspirin,
and
caffeine
in
alleviat-
ing
migraine
headache
pain:
three
double-blind,
randomized,
placebo-controlled
trials.
Arch
Neurol.
1998;55:210-217.
16.
Goldstein
J,
Hoffman
HD,
Armellino
JJ,
et
al.
Treat-
ment
of
severe,
disabling
migraine
attacks
in
an
over-
the-counter
population
of
migraine
sufferers:
results
from
three
randomized,
placebo-controlled
studies
of
the
combination
of
acetaminophen,
aspirin,
and
caf-
feine.
Cephalalgia.
1999;19:684-691.
17.
Ward
N,
Whitney
C,
Avery
D,
Dunner
D.
The
analgesic
effects
of
caffeine
in
headache.
Pain.
1991;44:151-155.
18.
Zeger
SL,
Liang
KY.
Longitudinal
data
analysis
for
discrete
and
continuous
outcomes.
Biometrics.
1986;42:121-130.
19.
Zuniga,
Phillips
CL,
Shugars
D,
et
al.
Analgesic
safety
and
efficacy
of
diclofenac
sodium
softgels
in
postop-
erative
dental
pain.
J
Oral
Maxillofac
Surg.
In
press.