Research

Original Investigation

Effect of Clopidogrel and Aspirin vs Aspirin Alone on Migraine

Headaches After Transcatheter Atrial Septal Defect Closure
The CANOA Randomized Clinical Trial
Josep Rodés-Cabau, MD; Eric Horlick, MD; Reda Ibrahim, MD; Asim N. Cheema, MD; Marino Labinaz, MD; Najaf Nadeem, MD; Mark Osten, MD;
Mélanie Côté, MSc; Josep Ramon Marsal, MSc; Donald Rivest, MD; Alier Marrero, MD; Christine Houde, MD

Supplemental content at
IMPORTANCE The occurrence of new-onset migraine attacks is a complication of jama.com
transcatheter atrial septal defect (ASD) closure. It has been suggested that clopidogrel may
reduce migraine attacks after ASD closure.

OBJECTIVE To assess the efficacy of clopidogrel, used in addition to taking aspirin, for the
prevention of migraine attacks following ASD closure.

DESIGN, SETTING, AND PARTICIPANTS Randomized, double-blind clinical trial performed in 6

university hospitals in Canada. Participants were 171 patients with an indication for ASD
closure and no history of migraine.

INTERVENTIONS Patients were randomized (1:1) to receive dual antiplatelet therapy

(aspirin + clopidogrel [the clopidogrel group], n = 84) vs single antiplatelet therapy
(aspirin + placebo [the placebo group], n = 87) for 3 months following transcatheter ASD
closure. The first patient was enrolled in December 2008, and the last follow-up was
completed in February 2015.

MAIN OUTCOMES AND MEASURES The primary efficacy outcome was the monthly number of
migraine days within the 3 months following ASD closure in the entire study population. The
incidence and severity of new-onset migraine attacks, as evaluated by the Migraine Disability
Assessment questionnaire, were prespecified secondary end points. A zero-inflated Poisson
regression model was used for data analysis.

RESULTS The mean (SD) age of the participants was 49 (15) years and 62% (106) were
women. Patients in the clopidogrel group had a reduced mean (SD) number of monthly
migraine days within the 3 months following the procedure (0.4 [95% CI, 0.07 to 0.69] days)
vs the placebo group (1.4 [95% CI, 0.54 to 2.26] days; difference, −1.02 days [95% CI, −1.94 to
−0.10 days]; incident risk ratio [IRR], 0.61 [95% CI, 0.41 to 0.91]; P = .04) and a lower
incidence of migraine attacks following ASD closure (9.5% for the clopidogrel group vs 21.8%
for the placebo group; difference, −12.3% [95% CI, −23% to −1.6%]; odds ratio [OR], 0.38
[95% CI, 0.15 to 0.89]; P = .03). Among patients with migraines, those in the clopidogrel
group had less-severe migraine attacks (zero patients with moderately or severely disabling
migraine attacks vs 37% [7 patients] in the placebo group; difference, −36.8% [95% CI,
−58.5% to −15.2%]; P = .046). There were no between-group differences in the rate of
patients with at least 1 adverse event (16.7% [14 patients] in the clopidogrel group vs 21.8%
[19 patients] in the placebo group; difference, −5.2% [95% CI, −17% to 6.6%]; P = .44).

CONCLUSIONS AND RELEVANCE Among patients who underwent transcatheter ASD closure,
the use of clopidogrel and aspirin, compared with aspirin alone, resulted in a lower monthly
frequency of migraine attacks over 3 months. Further studies are needed to assess
Author Affiliations: Author
generalizability and durability of this effect. affiliations are listed at the end of this
article.
TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT00799045 Corresponding Author: Josep
Rodés-Cabau, MD, Quebec Heart
and Lung Institute, Laval University,
JAMA. 2015;314(20):2147-2154. doi:10.1001/jama.2015.13919 Quebec City, QC G1V 4G5, Canada
Published online November 9, 2015. (josep.rodes@criucpq.ulaval.ca).

(Reprinted) 2147

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 08/05/2019

 Research Original Investigation
T
he occurrence of new-onset migraine attacks has been
reported in approximately 15% of patients following
transcatheter atrial septal defect (ASD) closure, with the
majority of initial episodes occurring within the days to weeks
following the procedure.1-10
Antithrombotic therapy following transcatheter ASD clo-
sure remains empirical, with aspirin for 6 months being com-
monly prescribed. Preliminary observational retrospective
studies have suggested an association with lower incidence and
severity of migraine headaches following ASD closure when
ticlopidine or clopidogrel is added to aspirin treatment.7-10 Sub-
sequently there has been a widespread use of dual antiplate-
let therapy following ASD closure, although no randomized
data have confirmed such observations. The objective of this
study was to assess the effect of clopidogrel on preventing the
occurrence and reducing the number of new-onset migraine
headache episodes following transcatheter ASD closure in pa-
tients with no history of migraine headaches.
Methods
Study Design and Patients
This was a randomized, double-blind clinical trial. CANOA
(Clopidogrel for the Prevention of New-Onset Migraine Head-
ache Following Transcatheter Closure of Atrial Septal De-
fects) was conducted in 6 centers in Canada, and it was ap-
proved by Health Canada and the local ethics committee of each
participating center. All patients provided written informed
consent for trial participation (trial protocol reported in
Supplement 1).
The study included patients aged 18 years and older with
a clinical indication for transcatheter ASD closure. Within the
60 days prior to undergoing the procedure, each patient com-
pleted a structured migraine headache questionnaire (eAp-
pendix 1 in Supplement 1), which was subsequently evalu-
ated by a neurologist to determine whether a history of
migraine headaches was present. Patients with a history of mi-
graine headaches, based on the International Headache Soci-
ety criteria,11 were excluded. Other exclusion criteria were al-
lergy or intolerance to any of the antithrombotic drugs used
in the study (aspirin, clopidogrel), need for anticoagulation
therapy, previous stroke, pregnancy or breastfeeding, not using
an effective method of birth control in premenopausal women,
use of ASD closure devices other than the Amplatzer Septal Oc-
cluder device (St Jude Medical), unsuccessful ASD closure (de-
fined as no device implanted at the time of the procedure), and
failure to provide informed consent.
The transcatheter ASD closure procedure has been de-
scribed in detail in prior studies.12 The procedure was per-
formed though transfemoral venous approach under guid-
ance of echocardiography (transesophageal or intracardiac).
Following the measurement of the ASD size by echocardiog-
raphy, a further assessment of the ASD size was performed by
inflation of a sizing balloon within the ASD defect, which de-
termined the size of the ASD closure device. Device implan-
tation was performed using current techniques. The proce-
dures were performed while the patient was under full
2148 JAMA November 24, 2015 Volume 314, Number 20 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/05/2019
Clopidogrel and Aspirin vs Aspirin Alone in Treating Migraine Headaches
anticoagulation receiving intravenous heparin. Transtho-
racic echocardiography was performed at hospital discharge
and at 3 months following the procedure.
Patients were randomized (1:1) before the ASD closure pro-
cedure to receive either aspirin (80 mg/d) plus placebo (the pla-
cebo group) or aspirin (80 mg/d) plus clopidogrel (75 mg/d) (the
clopidogrel group). Random block sizes were used to conceal
treatment allocation from the patients, and randomization was
stratified by clinical center. The treatment was initiated (with-
out loading dose) within 24 hours prior to ASD closure and con-
tinued for 3 months thereafter. A headache diary was given to
each patient at the time of randomization. Patients were asked
to indicate, on a daily basis, the occurrence of headache epi-
sodes and describe the characteristics, duration, severity, and
medication used for headache relief. The occurrence of head-
ache episodes was also assessed at 1- and 3-month follow-up
using a structured migraine headache questionnaire, which in-
cluded the Migraine Disability Assessment (MIDAS) question-
naire (grade I [score of 0-5] indicates little or no disability, grade
II [score, 6-10] mild disability, grade III [score, 11-20] moder-
ate disability, and grade IV [score ≥21] severe disability). The
questionnaires were provided by the study coordinator of each
participating center. All questionnaires and headache diaries
were evaluated by 2 neurologists blinded to procedural de-
tails and treatment allocation. If headache episodes oc-
curred, the diagnosis of migraine attacks was established ac-
cording to the International Headache Society criteria.11 Also,
migraine attacks were further classified as with or without aura.
Efficacy Outcomes
The primary efficacy outcome was the monthly number of
new-onset migraine attacks within the 3 months following ASD
closure in the entire study population. Secondary efficacy end
points were the incidence of new-onset migraine attacks, total
number of migraine attacks during the first month and 3-month
period following ASD closure (the entire population and pa-
tients with migraine attacks only), monthly number of new-
onset migraine attacks within 3 months following ASD clo-
sure in patients with migraine attacks only, severity of migraine
attacks as evaluated by the MIDAS questionnaire at 3-month
follow-up, and time to first migraine episode. Secondary safety
end points included the incidence of adverse events (re-
corded at 1- and 3-month follow-up) including death, tran-
sient ischemic attack, stroke, bleeding complications, and ad-
verse drug reactions.
There were no changes to methods or trial outcomes af-
ter the trial commenced.
Statistical Analyses
Based on previous studies,3,4,9 we anticipated an incidence of
15% of new-onset migraine attacks following ASD closure, with
a median (interquartile range [IQR]) number of 5 (IQR, 1-23) mi-
graine days per month within the 3 months following the pro-
cedure among patients with migraine attacks. The addition of
clopidogrel therapy was expected to reduce the incidence and
number of migraine headaches by at least 50%.3,7 Consider-
ing that a substantial number of patients were not expected
to have migraine attacks following ASD closure, the mean num-
jama.com
 Clopidogrel and Aspirin vs Aspirin Alone in Treating Migraine Headaches Original Investigation Research

ber of migraine days per month by patient (including those who

Figure 1. Flowchart of the CANOA Study Population
did not have migraine attacks) was modeled by a mixture of
zero-truncated Poisson distribution and zero-constant distri-
291 Patients assessed for eligibility
bution, which was calculated with a logistic model (SAS Proc
Genmod program).13 The power of the test was computed for 71 Excluded (history of migraines)
several sample sizes by Monte Carlo simulations, and a total
number of 70 patients per group were estimated to provide 220 Randomized
80% power to detect differences between groups. To account
for a potential dropout rate of approximately 10%, the sample
109 Randomized to receive aspirin 111 Randomized to receive aspirin
size was increased to 160 patients. + clopidogrel + placebo
Qualitative variables were expressed as percentages and 84 Received intervention as 87 Received intervention as
randomized randomized
quantitative variables as mean (SD) or median (IQR). Qualita- 25 Did not receive intervention 24 Did not receive intervention
tive variables were compared using the χ2 or Fisher exact test. 10 ASD closure unsuccessful 9 ASD closure unsuccessful
6 ASD closure not attempted 11 ASD closure not attempted
Comparison of numerical baseline and procedural variables and 9 Informed consent 4 Informed consent
migraine variables (number of migraine days) within the group withdrawn withdrawn

of patients with new-onset migraine attacks was performed

Follow-up
using the t test or Mann-Whitney rank-sum test depending on 8 Did not complete treatment
variable distribution. Comparison of the mean number of 5 Adverse events (treatment
stopped by investigator)
monthly migraine days and total number of migraine days at 3 Reasons unknown (treatment
1- and 3-month follow-up in the entire study population (with stopped by patient)
and without migraines) were performed using the zero-
inflated Poisson (ZIP) regression model. ZIP models provide 84 Included in primary analysis
Follow-up
8 Did not complete treatment
4 Adverse events (treatment
stopped by investigator)
4 Reasons unknown (treatment
stopped by patient)
87 Included in primary analysis

the methodology to explain the excess zeros by modeling the

data as a mixture of 2 separate distributions: first, a Poisson ASD indicates atrial septal defect.

distribution to model the counts among the patients who had

at least 1 migraine attack; and second, a logistic regression to
model the occurrence (or not) of new-onset migraine epi-
sodes. The effects estimated are incident risk ratios (IRRs) for
the Poisson distribution, which can be interpreted as the ra-
tio of expected mean number of migraine days per month dur-
ing the 3 months compared between the placebo and clopi-
dogrel groups and the odds ratio (OR) for the logistic model.
Additionally, the effect of the treatment on the new-onset mi-
graine attacks was estimated using logistic regression. For the
other 2 secondary end points (total migraine days at the first
month and at 3 months), a ZIP model was also used.
Excluding the patients nonadherent to allocated treat-
ment, the efficacy analyses were repeated. Differences were
considered statistically significant at P values of less than .05
(2-tailed). No type-I error from multiple comparisons of sec-
ondary outcomes was addressed in the analyses and the re-
sults from these analyses should be interpreted as explor-
atory. The data were analyzed using SAS version 9.1.3 and
RStudio version 0.98.953.
Results
The flow of patient participation through the trial is shown in
Figure 1. Patients were enrolled from December 2008 to No-
vember 2014, and the last patient follow-up was completed in
February 2015. A total of 291 patients diagnosed with an ASD
for whom transcatheter ASD closure was planned were
screened. Of these, 71 patients (24%) were excluded because
of a history of migraine attacks and were included in a pro-
spective migraine registry. Additionally, 49 patients were ex-
cluded at the time of the procedure because device implanta-
tion was not attempted (n = 17), was unsuccessful (n = 19), or
because the patient declined to continue in the study after ASD
closure (n = 13). The final study population consisted of 171 pa-
tients: 87 in the placebo group, and 84 patients in the clopi-
dogrel group. A total of 155 patients (91%) were adherent with
the treatment during the study period (>80% of tablets taken
as scheduled during the study period, as determined by tab-
let count). The treatment was temporarily or definitely stopped
by the investigator due to adverse events in 9 patients and by
the patients (for unknown reasons) in 7 patients.
The baseline and procedural characteristics of the study
population, according to treatment allocation, are reported in
Table 1. The mean (SD) age of the study population was 49 (15)
years, and 106 (62%) patients were women. The mean size of
the ASD was 16.3 (5.5) mm, as measured by transesophageal
echocardiography, and 20.9 (5.7) mm by balloon measure-
ment. The median device size was 22 (IQR, 18-26) mm. There
were no significant between-group differences regarding base-
line and procedural characteristics.
A total of 27 patients (15.8%) had the diagnosis of new-
onset migraine attacks within the 3 months following the pro-
cedure (14 [52%] of them with aura). In these patients, the me-
dian number of total migraine days was 12 (IQR, 6-20), and the
median monthly number of migraine days was 4 (IQR, 2-7). The
primary and secondary efficacy end points for the entire study
population are shown in Table 2. The mean (95% CI) number
of monthly migraine days (primary outcome) was lower in the
clopidogrel group (0.4 [95% CI, 0.07 to 0.69] days) than in the
placebo group (1.4 [95% CI, 0.54 to 2.26] days; difference, −1.02
days [95% CI, −1.94 to −0.10]; IRR, 0.61 [95% CI, 0.41 to 0.91];
P = .04). The incidence of new-onset migraine attacks was
lower in the clopidogrel group (9.5% [8] of patients) than in
the placebo group (21.8% [19] of patients; difference, −12.3%
[95% CI, −23% to −1.6%]; OR, 0.38 [95% CI, 0.15 to 0.89];

jama.com (Reprinted) JAMA November 24, 2015 Volume 314, Number 20 2149

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 08/05/2019

 Research Original Investigation Clopidogrel and Aspirin vs Aspirin Alone in Treating Migraine Headaches

tient in the placebo group received preventive medication (ami-

Table 1. Baseline and Procedural Characteristics of the Study Population
tripyline) during the study period. A total of 77 patients had
Treatment Group, No (%)a nonmigraine headaches within the 3 months following the
Aspirin + Clopidogrel Aspirin + Placebo transcatheter ASD closure procedure, 41 patients (49%) in the
(n = 84) (n = 87)
Age, mean (SD), y 49 (16) 48 (15)
aspirin + clopidogrel group, and 36 patients (41%) in the aspi-
rin group, P = .36.
Male sex 36 (42.9) 29 (33.3)
The main outcomes, excluding the nonadherent patients
Smoker 11 (12.9) 11 (13.1)
(ie, <80% of treatment tablets taken as scheduled during the
Hypertension 20 (23.8) 24 (27.6)
study period, n = 16), are shown in eTables 1 and 2 in
Diabetes 5 (5.9) 8 (9.2)
Supplement 2. A total of 155 patients were adherent with the
New York Heart Association class
treatment during the study period (76 in the clopidogrel group,
I 68 (82.9) 61 (73.5)
79 in the placebo group). The mean (SD) number of migraine
II 12 (14.6) 19 (22.9)
days per month within the 3 months following ASD closure was
III or IV 2 (2.4) 3 (3.6)
lower in the clopidogrel group (0.4 [1.4] days) than in the pla-
Pulmonary pressure, mean (SD), 20.4 (5.6) 22.4 (6.3) cebo group (1.6 [4.3] days; difference, −1.2 days [95% CI, −2.2
mm Hg
Qp/Qs ratio, mean (SD) 1.93 (0.84) 1.89 (0.81)
to −0.19]; IRR, 0.58 [95% CI, 0.38 to 0.90]; P = .02). Also, the
incidence of migraine attacks was lower in the clopidogrel
Atrial septal aneurysm 11 (15.5) 11 (16.4)
group (9.2% [7] of patients) than in the aspirin group (24.1%
ASD size, mean (SD), mm
[19] of patients; difference, −14.8% [95% CI, −26.3% to −3.4%];
Measured by TEE 15.7 (5.7) 16.9 (5.7)
OR, 0.32 [95% CI, 0.13 to 0.81]; P = .02).
Measured by balloon 20.3 (6.3) 21.7 (5.3)
The occurrence and type of adverse events during the trial,
Device size, median (IQR), mm 22 (18-26) 22 (19-28)
according to treatment allocation, are shown in Table 4. There
Hospitalization length, 1 (1-1) 1 (1-1)
median (IQR), d were no deaths or stroke events during the entire study pe-
Residual shunt 27 (32.1) 26 (29.8) riod. One patient in the placebo group experienced a tran-
(hospital discharge) sient ischmic attack. There were no between-group differ-
b
Mild 26 (30.1) 25 (28.7) ences in the occurrence of bleeding complications (no major
Moderate to severec 1 (1.2) 1 (1.1) bleeding events; and for minor bleeding, 5.9% in the clopido-
Residual shunt (3-mo follow-up) 8 (12.7) 9 (10.3) grel group vs 1.2% in the placebo group [difference, 4.8% {95%
Mildb 8 (12.7) 8 (9.2) CI, −0.7% to 10.3%}]; P = .11). The rate of patients with any ad-
Moderate to severec 0 1 (1.1) verse event throughout the study period was similar be-
Abbreviations: ASD, atrial septal defect; IQR, interquartile range; tween groups (16.7% [14 patients] in the clopidogrel group vs
Qp/Qs, pulmonary flow/systemic flow ratio; 21.8% [19 patients] in the placebo group [difference, −5.2%
TEE, transesophageal echocardiography. [95% CI, −17.0% to 6.6%]; P = .44).
a
Values are reported as No. (%) unless otherwise indicated.
b
Mild if color jet width (Doppler echocardiography) is 2 mm or less.14
c
Moderate to severe if color jet width (Dopler echocardiography) is greater
than 2 mm.14 Discussion
Dual antiplatelet therapy (aspirin + clopidogrel) following
P = .03. The secondary efficacy end points for the group of pa- transcatheter ASD closure was associated with a statistically
tients with migraine attacks following ASD closure (n = 27) are significant reduction in the occurrence and number of new-
shown in Table 3. In the presence of migraine attacks follow- onset migraine headaches within the 3 months following the
ing ASD closure, the addition of clopidogrel was associated with procedure. Among patients with migraine attacks, those on
a reduced migraine severity (as evaluated by the MIDAS ques- dual antiplatelet therapy tended to experience less-severe mi-
tionnaire [zero patients in the clopidogrel group had moder- graine attacks. No significant increase in adverse events was
ate or severe disabling migraine attacks vs 36.8% {7 patients} observed with the use of dual vs single antiplatelet therapy.
in the placebo group; difference, −36.8% {95% CI, −58.5% to Transcatheter ASD closure has been well established as the
−15.2%}]; P = .046). There were no between-group differ- treatment of choice for most patients with hemodynamically
ences regarding the type of migraine (aura vs no aura), mi- significant ASD. This treatment offers a very high success rate
graine duration, or time to first migraine episode. Figure 2 together with an extremely low rate of complications (includ-
shows the timing of the first and subsequent migraine at- ing cerebrovascular events), and the minimally invasive ap-
tacks during the 3 months following ASD closure according to proach, when compared with surgery, allows a much more
treatment allocation. In the clopidogrel group, 6 patients (75%) rapid recovery.14 This combination of high success rate, low
received paracetamol during the migraine episodes. In the as- rate of complications, and more rapid recovery rate is of par-
pirin group, 16 patients (84%) received analgesic therapy dur- ticular importance considering that most adult patients re-
ing the migraine attacks, distributed as follows: acetamino- ceiving this treatment are of the working age. However, a limi-
phen/paracetamol (10 patients), ibuprofen (4 patients), tation of transcatheter ASD closure is the occurrence of
ibuprofen + acetaminophen/paracetamol (1 patient), and apo- migraine attacks—affecting approximately 15% patients and ex-
tramadol + acetaminophen/paracetamol (1 patient). One pa- tending to approximately 1 week per month during the first

2150 JAMA November 24, 2015 Volume 314, Number 20 (Reprinted) jama.com

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 08/05/2019

 Clopidogrel and Aspirin vs Aspirin Alone in Treating Migraine Headaches Original Investigation Research

Table 2. Occurrence and Number of New-Onset Migraine Attacks Within the 3 Months Following ASD Closure

Treatment Group (95% CI)

Aspirin + Aspirin +
Clopidogrel Placebo Difference IRR OR
(n = 84) (n = 87) (95% CI) (95% CI)a P Value (95%CI)b P Value
Migraine days, mean (SD) [95% CI]
Days per mo (primary outcome) 0.4 (1.4) 1.4 (4.1) −1.02 0.61 .04 0.39 .02
[0.07 to 0.69] [0.54 to 2.26] (−1.94 to −0.10) (0.41 to 0.91) (0.16 to 0.95)
Days (first mo) 0.5 (2.2) 1.5 (4.5) −1.01 0.84 .34 0.33 .02
[0.02 to 0.96] [0.57 to 2.43] (−2.06 to 0.04) (0.59 to 1.20) (0.13 to 0.83)
Total migraine days at 3 mo 1.0 (4.1) 3.8 (10.6) −2.74 0.61 <.001 0.38 .03
[0.12 to 1.88] [1.5 to 5.95] (−5.14 to −0.34) (0.48 to 0.77) (0.16 to 0.92)
New-onset migraine attacks, No. (%) 8 (9.5) 19 (21.8) −12.3% 0.38 .03
(−23.0% to −1.6%) (0.15 to 0.89)
With aura 3 (37.5) 11 (57.9) −20.4% .33
(−60.6% to 19.8%)

Abbreviations: IRR, incidence risk ratio; OR, odds ratio.

a
Zero-inflated Poisson regression model (number of migraine attacks).
b
Zero-inflated Poisson regression model (probability of migraine attacks).

Table 3. Migraine Characteristics in Patients With Migraine Headaches

Treatment Group
Aspirin + Clopidogrel Aspirin + Placebo Difference
(n = 8) (n = 19) (95% CI) P Value
Migraine, median (IQR)
Days per mo 3.5 (2 to 5) 5 (2 to 8) −2.5 (−7.7 to 2.7) .39
Days (first mo) 3.5 (1 to 9) 4 (2 to 9) −1.8 (−7.9 to 4.2) .31
Days (3 mo) 7 (5 to 14.5) 13 (6 to 20) −6.8 (−20.1 to 6.5) .30
Migraine duration, median (IQR), 4 (3.5 to 5.5) 4 (3 to 6) −2.4 (−6.6 to 1.7) .94
hours per attack
Aura, No. (%) 3 (37.5) 11 (57.9) −20.4% (−60.6% to 19.8%) .33
Time to first migraine attack, 7 (5 to 15) 4 (2 to 15) 4.4 (−8.1 to 17.0) .42
median (IQR), days from procedure
MIDAS, No. (%)
I-II 8 (100) 12 (63.2) .046
−36.8% (−58.5% to −15.2%)
III-IVa 0 7 (36.8)

Abbreviations: IQR, interquartile range; MIDAS, Migraine Disability Assessment.

a
Moderate or severe disabling migraine attacks.

months following the procedure.1-10 This limitation repre-
sents a significant burden of medical consultations in addi-
tion to missed work or school days and reduced productivity.
The present study, which to our knowledge is the first ran-
domized trial in this field, confirmed prior observations on the
incidence of migraine attacks following transcatheter ASD clo-
sure and demonstrated the usefulness of dual antiplatelet
therapy for preventing and reducing, by more than 50%, the
burden of such migraine episodes following the procedure.
However, the reduction in monthly migraine days in the en-
tire study population (0.4 days in the clopidogrel broup vs 1.4
days in the placebo group) may appear modest due to the high
number of patients with no migraine headaches after the pro-
cedure. Moreover, the addition of clopidogrel therapy re-
duced the severity of migraine episodes, with no patient in the
dual antiplatelet therapy group presenting moderately or se-
verely disabling headache episodes compared with more than
one-third of the patients receiving single antiplatelet therapy
(aspirin only). No increase in adverse events was observed with
the addition of clopidogrel. Whether the optimal duration of
clopidogrel treatment should be 3 months or longer follow-
ing ASD closure will need to be determined in future studies.
Also, no loading dose of clopidogrel was used in this trial; the
potential additional effect of a loading dose of clopidogrel on
the prevention of migraine headaches following ASD closure
needs to be further evaluated.
An association between interatrial shunts and migraine has
been put forward in various observational studies,15-17 and a
significant reduction of migraine attacks following percuta-
neous shunt closure has also been suggested.15,16 The random-
ized trials evaluating the effects of atrial shunt closure on pa-
tients with migraine headaches have provided controversial
results, with negative or borderline primary outcomes and
some positive secondary outcomes like in the PRIMA trial.18,19
Although the pathophysiology linking interatrial shunts and
chronic migraine headaches is probably different from the one
linking ASD closure with new-onset migraines, the results of
this trial should be considered when designing or analyzing the
results of past, current, and future studies assessing the effi-
cacy of interatrial shunt closure for the treatment of mi-

jama.com (Reprinted) JAMA November 24, 2015 Volume 314, Number 20 2151

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 08/05/2019

 Research Original Investigation Clopidogrel and Aspirin vs Aspirin Alone in Treating Migraine Headaches

Figure 2. Timing of the First and Subsequent Migraine Attacks During the 3 Months Following ASD Closure

Aspirin + clopidogrel (8 individuals) Aspirin + clopidogrel

First migraine episode
Subsequent migraine episodes
Individual

Aspirin + placebo
First migraine episode
Subsequent migraine episodes

0 15 30 45 60 75 90
Days

Aspirin + placebo (19 individuals)

Individual

0 15 30 45 60 75 90 The first vertical column of data

Days markers on the left indicates day 1.
Day 0 was the day of ASD closure.

Table 4. Adverse Events at 3-Month Follow-up

Treatment Group, No. (%)

Aspirin + Clopidogrel Aspirin + Placebo Difference
Adverse Eventsa (n = 84) (n = 87) (95% CI) P Value
a
Zero patients in either group
Transient ischemic attack 0 1 (1.2) −1.1 (−3.4 to 1.1) >.99
experienced adverse events of
b
Minor bleeding 5 (5.9) 1 (1.2) 4.8 (−0.7 to 10.3) .11 death, pericardial effusion, device
Access site complications 0 1 (1.2) −1.1 (−3.4 to 1.1) >.99 embolization, need for cardiac
surgery, device thrombosis, or
Phlebitis 0 1 (1.2) −1.1 (−3.4 to 1.1) >.99
stroke during the study period.
Atrial fibrillation 2 (2.4) 5 (5.8) −3.4 (−9.2 to 2.5) .44 b
Minor adverse events associated
Palpitations 4 (4.8) 6 (6.9) −2.1 (−9.1 to 4.9) .75 with bleeding included epistaxis
Vertigo 0 2 (2.3) −2.3 (−5.4 to 0.9) .49 (n = 4), gingival bleeding (n = 1), and
minor hematuria (n = 1). There were
Cutaneous rash 3 (3.6) 2 (2.3) 1.3 (−3.8 to 6.3) .68
zero major bleeding adverse events.

graines. The addition of clopidogrel to the treatment of pa-
tients allocated to the transcatheter shunt closure group must
be considered as a potential confounding factor that may lead
to overestimating the real effect of shunt closure in such pa-
tients. Spencer et al20 showed a significant positive effect of
clopidogrel therapy on migraine patients with a patent fora-
men ovale before transcatheter closure. Migraine trials in this
setting should consider adjusting for clopidogrel therapy in case
of unbalanced treatment between groups.
The effects of a specific antithrombotic regimen, such as the
use of clopidogrel on the occurrence of migraine attacks, pro-
vide further insight into the potential role of platelet-coagulation
activation and prothrombotic status in the pathogenesis of mi-
graine. The association between migraine and ischemic stroke
has been well established,21,22 and several studies have linked
migraine attacks with abnormalities of platelet morphology and
function and also with coagulation disorders.23-26 Some stud-
ies have shown a reduction in migraine attacks with the use of
anticoagulation or antiplatelet (clopidogrel) therapy in specific
subsets of patients,20,26 but 2 randomized studies failed to dem-
onstrate a significant effect of warfarin or clopidogrel on migraine
attacks.27,28 The present study showed that in the presence of
a specific trigger, clopidogrel may be highly effective in the pre-
vention of migraine attacks. The occurrence of increased plate-
let aggregation following ASD closure has been demonstrated,29
and this may have increased the release of substances such as
serotonin or proinflammatory cytokines like interleukins 1, 6,
and 8 and tumor necrosis factor α, which have been linked to
migraine attacks.23-25
Apart from its antiplatelet effects, clopidogrel has also
proven antioxidant and anti-inflammatory properties.30 Sev-
eral studies have also shown improved endothelial function
and nitric oxide bioavailability with the use of clopidogrel.31-33
The vasoprotective effects through adenosine phosphate re-
ceptor blockade may also have played a role in the prevention
of migraine attacks.34 Further studies to evaluate the patho-
physiology linking clopidogrel with antimigraine effects are
warranted. Also, it remains to be elucidated whether the ef-

2152 JAMA November 24, 2015 Volume 314, Number 20 (Reprinted) jama.com

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 08/05/2019

 Clopidogrel and Aspirin vs Aspirin Alone in Treating Migraine Headaches Original Investigation Research

fects of clopidogrel on migraine attacks are a class effect or
specific to that medication.
Limitations
No platelet reactivity examinations were performed to evalu-
ate clopidogrel resistance,35 and thus it remains unknown
whether the patients with migraine attacks in the dual anti-
platelet group had some degree of clopidogrel resistance. These
results were obtained in patients undergoing ASD closure with
the Amplatzer Septal Occluder device and may not apply to
other transcatheter ASD closure devices. Migraine headaches
following ASD closure may decrease over time36; studies with
a longer-term follow-up are needed to determine the tran-
sient vs permanent behavior of new-onset migraine head-
aches after ASD closure.
Conclusions
Among patients who underwent transcatheter ASD closure,
the use of clopidogrel and aspirin, compared with aspirin
alone, resulted in a lower monthly frequency of migraine
attacks over 3 months. Further studies are needed to assess
generalizability and durability of this effect.

ARTICLE INFORMATION
Published Online: November 9, 2015.
doi:10.1001/jama.2015.13919.
Author Affiliations: Department of Cardiology,
Quebec Heart and Lung Institute, Quebec City,
Quebec, Canada (Rodés-Cabau, Côté); Department
of Cardiology, Toronto General Hospital, Toronto,
Ontario, Canada (Horlick, Osten); Department of
Cardiology, Montreal Heart Institute, Montreal,
Quebec, Canada (Ibrahim); Department of
Cardiology, St. Michael’s Hospital, Toronto, Ontario,
Canada (Cheema); Department of Cardiology,
Ottawa Heart Institute, Ottawa, Ontario, Canada
(Labinaz); Department of Cardiology, Queen
Elizabeth II Health Sciences Centre, Halifax, Nova
Scotia, Canada (Nadeem); Epidemiology Unit of the
Cardiology Department, Vall d'Hebron Hospital,
Universitat Autonoma de Barcelona, Barcelona,
Spain (Marsal); Department of Neurology, Hôtel
Dieu de Lévis, Quebec City, Quebec, Canada
(Rivest); Department of Neurology, Centre
Hospitalier Universitaire Georges. L Dumont,
Moncton, New Brunswick, Canada (Marrero);
Department of Pediatric Cardiology, Centre
Hospitalier Universitaire de Québec, Quebec City,
Quebec, Canada (Houde).
Author Contributions: Dr Rodés-Cabau and Ms
Côté had full access to all the data in the study and
take responsibility for the integrity of the data and
the accuracy of the data analysis.
Study concept and design: Rodés-Cabau.
Acquisition, analysis, or interpretation of data:
Rodés-Cabau, Horlick, Ibrahim, Cheema, Labinaz,
Nadeem, Osten, Coté, Marsal, Rivest, Marrero,
Houde.
Drafting of the manuscript: Rodés-Cabau, Coté,
Marsal.
Critical revision of the manuscript for important
intellectual content: Rodés-Cabau, Horlick, Ibrahim,
Cheema, Labinaz, Nadeem, Osten, Marsal, Rivest,
Marrero, Houde.
Statistical analysis: Coté, Marsal.
Obtained funding: Rodés-Cabau.
Administrative, technical, or material support:
Rodés-Cabau, Ibrahim, Cheema, Labinaz, Nadeem,
Osten, Marsal, Rivest.
Study supervision: Rodés-Cabau, Horlick, Ibrahim,
Nadeem, Osten, Marrero, Houde.
Conflict of Interest Disclosures: All authors have
completed and submitted the ICMJE Form for
Disclosure of Potential Conflicts of Interest. Dr
Horlick reports receipt of honoraria for lectures
outside the submitted work from St Jude Medical
and an institutional unrestricted grant for research,
education, and administrative support from St Jude
Medical. The other authors report no disclosures.
Funding/Support: This study was funded by
unrestricted grants from Sanofi and St Jude Medical
and a grant from the Foundation of the Quebec
Heart and Lung Institute.
Role of the Funder/Sponsor: There was no role of
any organization or sponsor in any of the following:
design and conduct of the study, collection,
management, analysis, and interpretation of the
data, preparation, review, and approval of the
manuscript, and decision to submit the manuscript
for publication.
REFERENCES
1. Rodés-Cabau J, Molina C, Serrano-Munuera C,
et al. Migraine with aura related to the
percutaneous closure of an atrial septal defect.
Catheter Cardiovasc Interv. 2003;60(4):540-542.
2. Fernández-Mayoralas DM, Fernández-Jaén A,
Muñoz-Jareño N, Gutiérrez-Larraya F, Calleja-Pérez
B, San Antonio Arce V. Migraine symptoms related
to the percutaneous closure of an ostium
secundum atrial septal defect: report of four
paediatric cases and review of the literature.
Cephalalgia. 2007;27(6):550-556.
3. Mortelmans K, Post M, Thijs V, Herroelen L,
Budts W. The influence of percutaneous atrial
septal defect closure on the occurrence of migraine.
Eur Heart J. 2005;26(15):1533-1537.
4. Yew G, Wilson NJ. Transcatheter atrial septal
defect closure with the Amplatzer septal occluder:
five-year follow-up. Catheter Cardiovasc Interv.
2005;64(2):193-196.
5. Wei SH, Fan PC, Kuo PH, Chiou LC, Wang JK.
Calcitonin gene-related peptide and size of the
atrial septal defect in new-onset migraine after
transcatheter closure: results of a preliminary study.
Headache. 2012;52(6):985-992.
6. Castellini P, Lambru G, Manzoni GC, Torelli P.
Atrial septal defect closure and migraine with aura:
is there a correlation? Neurol Sci. 2009;30(4):
339-342.
7. Sharifi M, Dehghani M, Mehdipour M,
Al-Bustami O, Emrani F, Burks J. Intense migraines
secondary to percutaneous closure of atrial septal
defects. J Interv Cardiol. 2005;18(3):181-183.
8. Wilmshurst PT, Nightingale S, Walsh KP,
Morrison WL. Clopidogrel reduces migraine with
aura after transcatheter closure of persistent
foramen ovale and atrial septal defects. Heart.
2005;91(9):1173-1175.
9. Rodés-Cabau J, Mineau S, Marrero A, et al.
Incidence, timing, and predictive factors of
new-onset migraine headache attack after
transcatheter closure of atrial septal defect or
patent foramen ovale. Am J Cardiol. 2008;101(5):
688-692.
10. Kato Y, Kobayashi T, Ishido H, Hayashi T,
Furuya D, Tanahashi N. Migraine attacks after
transcatheter closure of atrial septal defect.
Cephalalgia. 2013;33(15):1229-1237.
11. Headache Classification Subcommittee of the
International Headache Society. The International
Classification of Headache Disorders: 2nd edition.
Cephalalgia. 2004; 24(suppl 1): 9-160.
12. Kazmouz S, Kenny D, Cao QL, Kavinsky CJ,
Hijazi ZM. Transcatheter closure of secundum atrial
septal defects. J Invasive Cardiol. 2013;25(5):257-264.
13. Palta M. Quantitative Methods in Population
Health: Extensions of Ordinary Regression. Hoboken,
NJ: Wiley; 2003.
14. Du ZD, Hijazi ZM, Kleinman CS, Silverman NH,
Larntz K; Amplatzer Investigators. Comparison
between transcatheter and surgical closure of
secundum atrial septal defect in children and
adults: results of a multicenter nonrandomized trial.
J Am Coll Cardiol. 2002;39(11):1836-1844.
15. Lip PZ, Lip GY. Patent foramen ovale and
migraine attacks: a systematic review. Am J Med.
2014;127(5):411-420.
16. Luermans JG, Post MC, Temmerman F, et al.
Is a predominant left-to-right shunt associated with
migraine?: a prospective atrial septal defect closure
study. Catheter Cardiovasc Interv. 2009;74(7):
1078-1084.
17. Kato Y, Hayashi T, Kobayashi T, Tanahashi N.
Migraine prevalence in patients with atrial septal
defect. J Headache Pain. 2013;14:63.
18. Dowson A, Mullen MJ, Peatfield R, et al.
Migraine Intervention With STARFlex Technology
(MIST) trial: a prospective, multicenter,
double-blind, sham-controlled trial to evaluate the
effectiveness of patent foramen ovale closure with
STARFlex septal repair implant to resolve refractory
migraine headache. Circulation. 2008;117(11):
1397-1404.
19. Hildick-Smith D. Percutaneous closure of patent
foramen ovale in migraine with aura—a randomized
prospective study (PRIMA trial). Transcatheter
Cardiovascular Therapeutics Meeting, Washington,
DC, March 2014. http://www.sac.org.ar/wp-content
/uploads/2014/09/tct-2014-prima-a-prospective
-randomized-trial.pdf. Accessed October, 2, 2015.

jama.com (Reprinted) JAMA November 24, 2015 Volume 314, Number 20 2153

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ on 08/05/2019

 Research Original Investigation
20. Spencer BT, Qureshi Y, Sommer RJ.
A retrospective review of clopidogrel as primary
therapy for migraineurs with right to left shunt
lesions. Cephalalgia. 2014;34(11):933-937.
21. Kurth T, Diener HC. Migraine and stroke:
perspectives for stroke physicians. Stroke. 2012;43
(12):3421-3426.
22. Mawet J, Kurth T, Ayata C. Migraine and stroke:
in search of shared mechanisms. Cephalalgia.
2015;35(2):165-181.
23. Borgdorff P, Tangelder GJ. Migraine: possible
role of shear-induced platelet aggregation with
serotonin release. Headache. 2012;52(8):1298-1318.
24. Sarchielli P, Alberti A, Coppola F, et al.
Platelet-activating factor (PAF) in internal jugular
venous blood of migraine without aura patients
assessed during migraine attacks. Cephalalgia.
2004;24(8):623-630.
25. Danese E, Montagnana M, Lippi G. Platelets
and migraine. Thromb Res. 2014;134(1):17-22.
26. Maggioni F, Bruno M, Mainardi F, Lisotto C,
Zanchin G. Migraine responsive to warfarin: an
update on anticoagulant possible role in migraine
prophylaxis. Neurol Sci. 2012;33(6):1447-1449.
2154 JAMA November 24, 2015 Volume 314, Number 20 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/05/2019
Clopidogrel and Aspirin vs Aspirin Alone in Treating Migraine Headaches
27. Wammes-van der Heijden EA, Smidt MH,
Tijssen CC, van’t Hoff AR, Lenderink AW, Egberts
AC. Effect of low-intensity acenocoumarol on
frequency and severity of migraine attacks.
Headache. 2005;45(2):137-143.
28. Chambers JB, Seed PT, Ridsdale L. Clopidogrel
as prophylactic treatment for migraine: a pilot
randomised, controlled study. Cephalalgia. 2014;34
(14):1163-1168.
29. Pan G, Xie ZF, Zhang Y, Long SC, Xu XP,
Zhang ZW. Platelet activation through the efficacy
of aspirin in congenital heart disease patients
undergoing transcatheter closure of atrial septal
defects or ventricular septal defects. Genet Test Mol
Biomarkers. 2014;18(12):832-838.
30. Heitzer T, Rudolph V, Schwedhelm E, et al.
Clopidogrel improves systemic endothelial nitric
oxide bioavailability in patients with coronary artery
disease: evidence for antioxidant and
antiinflammatory effects. Arterioscler Thromb Vasc
Biol. 2006;26(7):1648-1652.
31. Warnholtz A, Ostad MA, Velich N, et al. A single
loading dose of clopidogrel causes dose-dependent
improvement of endothelial dysfunction in patients
with stable coronary artery disease: results of a
double-blind, randomized study. Atherosclerosis.
2008;196(2):689-695.
32. Giachini FR, Leite R, Osmond DA, et al.
Anti-platelet therapy with clopidogrel prevents
endothelial dysfunction and vascular remodeling in
aortas from hypertensive rats. PLoS One. 2014;9(3):
e91890.
33. Zhang YZ, Chen BL, Zhang W, Cao X.
Non-antiplatelet effect of clopidogrel: improving
endothelial function in Chinese healthy subjects
with different CYP2C19 genotype. Clin Exp
Pharmacol Physiol. 2015;42(1):22-26.
34. Tietjen GE. The role of the endothelium in
migraine. Cephalalgia. 2011;31(6):645-647.
35. Linden MD, Tran H, Woods R, Tonkin A.
High platelet reactivity and antiplatelet therapy
resistance. Semin Thromb Hemost. 2012;38(2):
200-212.
36. Voet A, Luermans JG, Thijs V, et al. New-onset
and persistent migraine early after percutaneous
atrial septal defect closure disappear at follow-up.
Acta Clin Belg. 2008; 63(4):262-268.
jama.com