e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 1 7 ( 2 0 1 3 ) 4 5e4 9

Official Journal of the European Paediatric Neurology Society

Original article

Comparison of flunarizine and topiramate for the prophylaxisof pediatric migraines

Hunmin Kima, Sung Hwan Byun b, Jon Soo Kim a, Byung Chan Lim b, Jong-Hee Chae b,Jieun Choi c, Ki Joong Kim b, Yong Seung Hwang b, Hee Hwang a,*aDepartment of Pediatrics, Seoul National University Bundang Hospital, Gyeonggi-do, Republic of KoreabDepartment of Pediatrics and Pediatric Clinical Neuroscience Center, Seoul National University Children’s Hospital, Seoul, Republic of KoreacDepartment of Pediatrics, Seoul National University Boramae Hospital, Seoul, Republic of Korea

a r t i c l e i n f o

Article history:

Received 8 August 2012

Received in revised form

25 September 2012

Accepted 6 October 2012

Keywords:

Topiramate

Flunarizine

Migraine prophylaxis

Pediatric

* Corresponding author. Division of PediatriGumi-ro, Bundang-gu, Seongnam-si, Gyeong

E-mail address: neuroandy@snubh.org (H1090-3798/$ e see front matter ª 2012 Europhttp://dx.doi.org/10.1016/j.ejpn.2012.10.001

a b s t r a c t

The purpose of this study was to compare the efficacy and tolerability of topiramate and

flunarizine for the prophylaxis of pediatricmigraines. A retrospectivemedical-record review

of patientswhounderwent prophylaxis after receiving a diagnosis ofmigrainewith aura and

without aura was performed. Only patients who completed at least 3 months of treatment

were included in the analysis. Response to treatment was assessed as the total number of

headache days/month. Patients with more than 50% reduction in headache days/month

were classified as responders. Responder rate, retention rate, and adverse-event rates were

also calculated from all patients who started on the prophylaxis. Further analyses were

performed using different patient groups with a cut-off age of 12 years. The responder rate

was 80% (89/111 patients) for flunarizine and 81% (122/150 patients) for topiramate, based on

a comparison among 261 patients. The retention rate was 67% for flunarizine and 63% for

topiramate and the adverse-event rate was 6% for flunarizine and 10% for topiramate. The

responder rate, the retention rate, and the adverse-event ratewere not significantly different

between flunarizine and topiramate. These findings were concordant between the preado-

lescent (6e12 years old) and adolescent (13e18 years old) groups. The efficacy and tolerability

of topiramate were not inferior to those of flunarizine for the prophylaxis of pediatric

migraines. These findings were observed in preadolescent and adolescent patients.

ª 2012 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights

reserved.

1. Introduction design.1e4 It is also common in Korean pediatric populations,

Migraines are a common neurological disorder that can

significantly impair the quality of life of pediatric patients. The

prevalence of migraines in the pediatric population ranges

from 3 to 15%, depending on the study population and

c Neurology, Departmengi-do 463-707, Republic o. Hwang).ean Paediatric Neurology

with an estimated prevalence of 8.7%.5 Patients with

migraines commonly experience worsening of headaches

with routine daily activities and the migraine headaches

themselves significantly interfere with the daily activities of

patients. Impairment of quality of life, such as absence from

t of Pediatrics, Seoul National University Bundang Hospital, 166f Korea. Tel.: þ82 31 787 7284; fax: þ82 31 787 4054.

Society. Published by Elsevier Ltd. All rights reserved.

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 1 7 ( 2 0 1 3 ) 4 5e4 946

school, is more common in patients with frequent or severe

migraine attacks. Indication for prophylactic treatment was

proposed in previous studies, which usually included patients

with frequentmigraine attacks (more than one or two attacks/

week) or severe disabling attacks.6,7

Preventive treatment with flunarizine (FNZ) is safe and

effective in pediatric patients with migraine8 and was rec-

ommended as a probable effective treatment in the American

Academy of Neurology (AAN) practice parameter in 2004.9

Randomized, double-blind, placebo-controlled trials of top-

iramate (TPM) also resulted in successful prevention of pedi-

atric migraines.10,11 However, no study has compared TPM

with FNZ as a prophylactic treatment of pediatric migraines in

real clinical practice settings. The purpose of this study was to

perform a comparison of TPM and FNZ in a large number of

patients with pediatric migraines from a single center. A

comparison of efficacy and tolerability was also performed in

different age subgroups of patients with migraine (preado-

lescents and adolescents).

Fig 1 e Age distribution of patients in the FNZ and TPM

group.

2. Materials and methods

This study was approved by the Institutional Review Board of

the Seoul National University Hospital. An electronic medical

record database search was performed using a clinical data

warehouse program. Patients diagnosed with migraines and

treated with either TPM or FNZ at the Seoul National Univer-

sity Children’s Hospital (SNUCH) and Seoul National Univer-

sity Bundang Hospital (SNUBH) from January 2005 to

December 2011were identified. The diagnosis ofmigrainewas

established based on the International Headache Society 2004

Classification12 after detailed history taking of headache

characteristics, disease course, and physical and neurological

examinations by pediatric neurologists (HH and YSH). Indi-

cations for prophylactic treatment included frequent head-

aches (more than 10 headache days/month) and frequent

prolonged migraine attacks that severely limited daily activi-

ties, such as school attendance. Selection and dose adjust-

ment of the specific prophylactic agent were made based on

the discretion of the treating physician. After the initiation of

prophylactic treatment, patients were evaluated monthly at

the pediatric neurology clinic. They were told to keep a head-

ache diary, and the response to treatment was assessed based

on the diary regarding headache days/month. Patients were

also told to record any adverse events, which were also

assessed at the follow-up visit. The starting dose of TPM was

1 mg/kg/day, taken once and at night. An increase in the

dosage was decided after at least 1 month of treatment. The

starting dose of FNZ was 5 mg/day, once and at night, and the

decision to increase the dose was also made after 1 month of

treatment.

After the database search, the medical records were

reviewed and only the patients with migraines with aura or

migraines without aura were included in the analysis. Clinical

information regarding sex, age at onset, age at diagnosis and

treatment, pre- and post-treatment headache days/month,

decrement of headache days after the treatment for each

patient, duration of treatment, reasons for withdrawal, and

any adverse events were retrieved. Detailed information

regarding continuation or discontinuation of treatment was

retrieved to analyze the outcome of initiation of the prophy-

lactic treatment in all patients. The outcomes of the initiation

of treatment were classified as: (1) continuation of the medi-

cation for longer than 3 months, (2) self-withdrawal because

of drug inefficacy, (3) withdrawal because of adverse events,

and (4) withdrawal without any specific reasons.

To maximize the comparability among the different

treatment groups, only the patients who completed at least 3

months of treatment with good compliance and had a period

of at least 1 month of post-treatment outcome evaluation

were included in the comparison. Response to prophylaxis

was categorized as headache free,�50% decrease in headache

days/month, <50% decrease in monthly headache days, and

no improvement. Patients who were headache free and with

a �50% decrease in headache days/month were considered as

responders and the remaining patients were classified as

nonresponders for statistical analysis. The rate of adverse

events and their profile were also compared. To identify

differences in efficacy and tolerability in the different age

groups, an additional analysis was performed after dividing

the patients according to age at the time of treatment (6e12

years for the preadolescent group and 13e18 years for the

adolescent group).

Statistical analysis was performed using SPSS 18.0 for

Windows. To compare patient characteristics, Student’s t test

was used for continuous variables. Pearson’s c2 test and

Fisher’s exact test were used for the analysis of discrete

variables based on sample size. Statistical significance was set

at P < 0.05.

3. Results

A total of 475 pediatric patients (206 males and 269 females)

who were treated prophylactically with either FNZ or TPM at

SNUCH and SNUBH were identified after a search and review

of electronic medical records. FNZ was prescribed to 212

patients (99 males and 113 females) and TPM was prescribed

to 263 patients (107 males and 156 females). The distribution

of each prophylactic agent in each age group is shown in Fig. 1.

Forty-six patients from the FNZ group and 24 patients from

the TPM group were lost in the follow-up. In the FNZ group,

67% (111/166) of patients were on medication longer than 3

Table 1 e Clinical characteristics and pre- and post-treatment headache frequencies of patientswith FNZ andTPM treatment, and comparison of FNZ and TPMtreatment after 3 months.

FNZ(n ¼ 111)

TPM(n ¼ 150)

P value

Sex (male:female) 42:69 64:86

Onset age 9.7 � 3.3 10.3 � 3.1 0.150

Age at diagnosis 11.4 � 3.0 11.6 � 3.0 0.613

Age at treatment 11.5 � 3.0 11.9 � 3.0 0.327

Duration of treatment 5.8 � 3.6 6.5 � 3.0 0.096

Pre-treatment frequency 10.7 � 6.2 11.9 � 5.9 0.116

Post-treatment frequency 2.9 � 4.3 3.1 � 4.7 0.756

Decreased HA-days/month 7.9 � 6.8 8.9 � 6.4 0.221

Adverse events 6/111 (5%) 15/150 (10%) 0.177

Response to treatment

Responder rate 80% (89/111) 81% (122/150) 0.710

HA free 38% (42/111) 45% (68/150)

�50% decrease 42% (47/111) 36% (54/150)

<50% decrease 5% (5/111) 5% (8/150)

No change 15% (17/111) 14% (20/150)

HA, headache; FNZ, flunarizine; TPM, topiramate.

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 1 7 ( 2 0 1 3 ) 4 5e4 9 47

months, vs 63% (150/239) in the TPM group. The final main-

tenance dose of FNZ was 5 mg/day (86/111 patients) or 10 mg/

day (25/111 patients). The maintenance dose of TPM ranged

from 25 to 100 mg/day and the final dose was 25 mg (24/150

patients), 50 mg (121/150 patients), or 100 mg (5/150 patients).

The reasons for withdrawal before 3 months of treatment

were low efficacy (29%), adverse events (2%), and unknown

reasons (2%) in the FNZ group and low efficacy (32%), adverse

events (3%), and unknown reasons (2%) in the TPM group

(Fig. 2).

Two hundred sixty-one patients (106 males and 155

females)were included in the comparison (111patients treated

with FNZ and 150 patients treated with TPM). Themean age of

onset, age at diagnosis, and age at treatment did not differ

significantly between the two groups. Therewas no significant

difference in pre- and post-treatment headache frequency

between the two groups. Treatment duration was shorter for

the FNZ group (5.8 months vs 6.5 months); however, this

difference was not significant. After 3 months of treatment,

38% of the patients in the FNZ group and 45% of the patients in

the TPMgroup reported freedom fromheadache. A decrease of

more than 50% in headache days/month occurred in 42% of

patients in the FNZgroupand36%ofpatients in theTPMgroup.

The responder rates were 80% for FNZ and 81% for TPM. There

was no significant difference in the responder rates between

the two groups (Table 1). Therewas no significant difference in

pre- and post-treatment headache frequency and in decrease

in headache days/month between the two groups. Adverse

events that did not lead to withdrawal were present in 5% of

patients in the FNZgroupand10%ofpatients in theTPMgroup.

Response to treatment was not significantly different

between the preadolescent and adolescent patient subgroups

and the overall patient group. The responder rates were not

significantly different between FNZ and TPM in the preado-

lescent and adolescent patient groups (Table 2). Male to

female ratio was 75:72 in preadolescent group was 38:76 in

adolescent group. The final maintenance doses of FNZ and

TPM in each of the two age groups are shown in Table 2.

Adverse events were present in 6% (10/166) of the patients

in the FNZ group and in 10% (23/229) of the patients in the TPM

group. The adverse-event rates were not significantly

different between the two groups. Common adverse events

were weight gain, drowsiness, and dizziness in the FNZ group,

Fig 2 e Diagram showing the distribution and outcome

and drowsiness, paresthesia, memory or language decline,

and anorexia in the TPM group (Table 3). Withdrawal due to

adverse events occurred in 4 patients with FNZ and 8 patients

with TPM. Symptoms were either transient or mild in other

patients that continued on prophylaxis.

We attempted to calculate the actual retention rates and

responder rates in all the patients that were initially treated

with the prophylactic agents, to compare these data to those

of intention-to-treat (ITT) analyses performed in randomized

controlled trials. The retention rate was 52% (111/212 patients)

for FNZ and 57% (150/263 patients) for TPM. The responder

rate was 42% (89/212 patients) for FNZ and 46% (122/262

patients) for TPM. The retention and responder rates of the

two groups did not differ significantly.

4. Discussion

FNZ was the only medication that was identified as a probable

effective treatment forpreventingpediatricmigraineheadaches

s after the prophylactic treatment, F/U, follow-up.

Table 2 e Clinical characteristics and pre- and post-treatment headache frequencies of patients with FNZ and TPMtreatment and comparison of FNZ and TPM treatment after 3 months sub-grouped by age at treatment of 12 years.

6w12 years old (n ¼ 147) 13w18 years old (n ¼ 114)

FNZ (n ¼ 64) TPM (n ¼ 83) FNZ (n ¼ 47) TPM (n ¼ 67)

Onset age 7.8 � 2.3 8.3 � 2.1 12.3 � 2.5 12.7 � 2.3

Age at diagnosis 9.3 � 2.0 9.5 � 1.9 14.3 � 1.5 14.3 � 2.0

Age at treatment 9.5 � 2.0 9.7 � 1.7 14.5 � 1.4 14.7 � 1.6

Duration of treatment 5.9 � 2.8 6.2 � 3.1 5.6 � 3.4 6.7 � 3.0

Pre-treatment frequency 10.5 � 6.4 11.4 � 6.0 11.0 � 6.1 12.5 � 5.7

Post-treatment frequency 2.7 � 4.4 2.1 � 3.4 3.1 � 4.2 4.2 � 5.7

Decreased HA-days/month 7.8 � 6.7 9.3 � 6.4 7.9 � 7.0 8.3 � 6.3

Adverse events 6% (4) 10% (8) 4% (2) 10% (7)

Final dosage (mg) (number) 5 (57) 25 (19) 5 (29) 25 (5)

10 (7) 50 (64) 10 (18) 50 (57)

100 (5)

Response to treatment

HA Free 39% (25) 47% (39) 36% (17) 43% (29)

>50% decrease 39% (25) 42% (35) 47% (22) 28% (19)

<50% decrease 6% (4) 1% (1) 2% (1) 10% (7)

No change 16% (10) 10% (8) 15% (7) 19% (12)

HA, headache; FNZ, flunarizine; TPM, topiramate.

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 1 7 ( 2 0 1 3 ) 4 5e4 948

in a systematic review from the AAN practice parameter.9

Similarly, FNZ was recognized as an effective prophylactic

treatment by the Cochrane Database Systemic Review, with

good-quality evidence.11 A double-blind, placebo-controlled,

crossover trial revealed that headache frequency and average

duration were significantly lower in patients during the FNZ

treatment period.8 Other studies that evaluated the efficacy of

FNZ showed: a significant reduction in headache frequency

(66%) and duration (51%),13 8/12 patients with a 75e100%

reduction in headache frequency,14 and 65% of patients with

a reduction in headache frequency of more than 50%.15 There

wasonlyoneclass IVstudy included in theanalysis that showed

a significant decrease in headache frequency (16.5 to 11.6

headaches/month, P < 0.001) and it was considered that those

data were insufficient to assess the evidence.9 The number of

studies that assessed the efficacy and tolerability of TPM has

increased. Headache frequency decreased significantly from

16.1 to 4.3 headaches/month16 and from 15.3 to 4.4 headaches/

month17 after prophylactic treatment with TPM. A reduction in

headachefrequencyofmore than50%occurred in28/37patients

Table 3 e Adverse events and frequency in all patientsprescribed with FNZ and TPM.

FNZ (n ¼ 10)a TPM (n ¼ 23)a

Drowsiness 4 8

Dizziness 3 3

Weight gain 5 0

Anorexia 0 4

Memory or language dysfunction 2 6

Paresthesia 0 6

Nervousness 0 1

Abdominal pain 0 1

Constipation 1 0

a Numbers represent total number of patients that reported

adverse events.

(76%).18 Two randomized, double-blind, placebo-controlled

studies using TPM reported conflicting results.10,19 Although

reduction in mean migraine days was reported in patients

receiving TPM treatment, these results were not significant

comparedwith those of the placebo group. A reduction inmean

migraine days ofmore than 75% occurred in 32% of the patients

treated with TPM, which was significantly more frequent than

what was observed in the placebo group.10 In another study of

adolescent patients, the responder rates were 83% for TPM

administered at 100 mg/day and 45% for the placebo; this

differencewas significant.19 Even though the number of studies

that report the efficacy and tolerability of TPM has increased,

these findings remain debatable because of the conflicting

results obtained in randomized controlled trials. In this study,

we compared the efficacy and tolerability of TPM and FNZ in

a real clinical setting of pediatric migraine prophylaxis.

The comparison of TPM and FNZ for migraine prophylaxis

in pediatric patients revealed that the efficacy and tolerability

of TPMwere not different from those of FNZ. Responder rates,

change in headache days/month, and adverse-event rates

were not significantly different between groups. Responder

rates of 80% for TPM and 81% for FNZ were similar to those of

previous studies that evaluated drug efficacy.14,15,18 The

responder rate of TPM that we estimated from all the patients

those were put on the prophylactic treatment as in ITT anal-

ysis of randomized controlled trials was similar to that of

previous report which showed 69.4% of patients showedmore

than 50% reduction.10 A recent study reported a similar effi-

cacy of TPM and FNZ for the prophylaxis of migraines in adult

patients.20 Our findings are similar to the responder rates of

66.7% in the FNZ group and 72.7% in the TPM group found in

that study. Both treatment groups showed similar tolerability,

including adverse events and retention. Withdrawal before 3

months of treatmentwas similar in both groups. Therewas no

significant difference in adverse-event rates estimated in all

patients with the decision and initiation of prophylaxis

included in the comparison. These results allowed us to infer

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 1 7 ( 2 0 1 3 ) 4 5e4 9 49

that TPM is as efficacious and tolerable as FNZ in the

prophylactic treatment of pediatricmigraine. The comparison

performed between patient subgroups according to age

revealed that TPM was as effective as FNZ in both the pread-

olescent and adolescent patient groups. These findings

suggest that the efficacy and tolerability of TPM are similar to

those of FNZ, even in patients younger than 12 years of age.

The common adverse events of TPM were drowsiness,

decline in memory or language function, and paresthesia,

whereas weight gain, drowsiness, and dizziness were the

adverse events of FNZ. Although the actual adverse-event

rates were higher in patients receiving TPM treatment, this

difference was not significant. Moreover, withdrawal because

of adverse events was not significantly different between the

two groups. The adverse-event rates found here were lower

than those of the randomized controlled trials, which re-

ported a 15e25% incidence of adverse events.10,16,18e20 One

explanation for this low adverse-event rate can be the

recording of self-reported adverse events instead of the use of

a questionnaire that lists the possible adverse events. With

the exception of the patients who withdrew from treatment

because of adverse events, all patients that were included in

the comparison continued taking medication because the

adverse events were tolerable. This finding was also similar in

both treatment groups. The possibility of weight gain or loss

should be considered in the decision of prophylactic treat-

ment and dizziness and possible decline in memory or

language function should be informed to the patients and

parents, before the initiation of prophylaxis.

This was a single-center study that used a large study

population. All patients had frequent migraine attacks or

severe attacks that significantly compromised their quality of

life. To maximize the comparability, we followed the stan-

dardized diagnostic criteria and achieved a well-documented

history of the headaches. A treatment period of at least 3

months was sufficient to assess the effects of prophylaxis.

These conditions were recommended by the Cochrane Data-

base Systemic Review as implications for future research.11

However, the retrospective design of our study was one of

the limitations of this analysis. In addition, even though our

headache clinic serves as a primary care center for patients

with migraines in the Korean health care system, a referral

bias may exist because the Seoul National University Hospital

is a tertiary care center. A longer observation period is neces-

sary for the assessment of long-term outcomes, and informa-

tion regarding the period after discontinuation of prophylactic

treatment should also be sought in future studies.

To the best of our knowledge, this is the first study that

compared TPM and FNZ for the prophylactic treatment of

pediatric migraines. The efficacy and tolerability of TPM were

not inferior to those of FNZ. The efficacy and tolerability of

TPM and FNZ were satisfactory and similar to those observed

in previous studies. These findings were also observed in

pediatric patients younger than 12 years of age.

Acknowledgments

This work was supported by the Industrial Strategic Tech-

nology Development Program, 10038690, Global Healthcare

Software Framework Development funded by the Ministry of

Knowledge Economy (MKE, Korea).

r e f e r e n c e s

1. Deubner DC. An epidemiologic study of migraine andheadache in 10e20 year olds. Headache 1977;17:173e80.

2. Lee LH, Olness KN. Clinical and demographic characteristicsof migraine in urban children. Headache 1997;37:269e76.

3. Zwart JA, Dyb G, Holmen TL, Stovner LJ, Sand T. Theprevalence of migraine and tension-type headaches amongadolescents in Norway. The Nord-Trondelag Health Study(Head-HUNT-Youth), a large population-basedepidemiological study. Cephalalgia 2004;24:373e9.

4. Sillanpaa M. Changes in the prevalence of migraine and otherheadaches during the first seven school years. Headache1983;23:15e9.

5. Rho YI, Chung HJ, Lee KH, et al. Prevalence and clinicalcharacteristics of primary headaches among school childrenin South Korea: a nationwide survey. Headache 2012;52:592e9.

6. Silberstein SD. Preventive migraine treatment. Neurol Clin2009;27:429e43.

7. Hershey AD. Current approaches to the diagnosis andmanagement of paediatric migraine. Lancet Neurol2010;9:190e204.

8. Sorge F, De Simone R, Marano E, et al. Flunarizine inprophylaxis of childhood migraine. A double-blind, placebo-controlled, crossover study. Cephalalgia 1988;8:1e6.

9. Lewis D, Ashwal S, Hershey A, et al. Practice parameter:pharmacological treatment of migraine headache in childrenandadolescents: report of theAmericanAcademyofNeurologyQuality Standards Subcommittee and the Practice Committeeof the Child Neurology Society. Neurology 2004;63:2215e24.

10. Winner P, Pearlman EM, Linder SL, et al. Topiramate formigraine prevention in children: a randomized, double-blind,placebo-controlled trial. Headache 2005;45:1304e12.

11. Victor S, Ryan SW. Drugs for preventing migraine headachesin children. Cochrane Database Syst Rev 2003;(4). CD002761,http://dx.doi.org/10.1002/14651858.CD002761.

12. Headache Classification Subcommittee of the InternationalHeadache Society. International classification of headachedisorders, 2nd edition. Cephalalgia 2004;24(Suppl. 1):8e152.

13. Sorge F, Marano E. Flunarizine v. placebo in childhoodmigraine. A double-blind study. Cephalalgia 1985;5:145e8.

14. Guidetti V, Moscato D, Ottaviano S, Fiorentino D, Fornara R.Flunarizine and migraine in childhood. An evaluation ofendocrine function. Cephalalgia 1987;7:263e6.

15. Visudtibhan A, Lusawat A, Chiemchanya S, Visudhiphan P.Flunarizine for prophylactic treatment of childhood migraine.J Med Assoc Thai 2004;87:1466e70.

16. Lakshmi CV, Singhi P, Malhi P, Ray M. Topiramate in theprophylaxis of pediatric migraine: a double-blind placebo-controlled trial. J Child Neurol 2007;22:829e35.

17. Unalp A, Uran N, Ozturk A. Comparison of the effectivenessof topiramate and sodium valproate in pediatric migraine. JChild Neurol 2008;23:1377e81.

18. Cruz MJ, Valencia I, Legido A, et al. Efficacy and tolerability oftopiramate inpediatricmigraine. PediatrNeurol2009;41:167e70.

19. Lewis D, Winner P, Saper J, et al. Randomized, double-blind,placebo-controlled study to evaluate the efficacy and safety oftopiramate for migraine prevention in pediatric subjects 12 to17 years of age. Pediatrics 2009;123:924e34.

20. Luo N, Di W, Zhang A, et al. A randomized, one-year clinicaltrial comparing the efficacy of topiramate, flunarizine, anda combination of flunarizine and topiramate in migraineprophylaxis. Pain Med 2012;13:80e6.