Headache ISSN 0017-8748C© 2006 by American Headache Society doi: 10.1111/j.1526-4610.2006.00599.xPublished by Blackwell Publishing

Research Submission

Low-Dose Topiramate Versus Lamotrigine in MigraineProphylaxis (The Lotolamp Study)

Praveen Gupta, MD; Sumit Singh, MD; Vinay Goyal, MD; Garima Shukla, MD;Madhuri Behari, MD

Objective.—To assess the efficacy and safety of topiramate and lamotrigine for prophylaxis in patients withfrequent migraine as compared to each other and to placebo.

Methods.—Sixty patients with frequent migraine (more than 4 attacks per month) from the headache clinic ata tertiary referral centre in India were randomized to receive 50 mg topiramate/lamotrigine or matching placebofor 1 month each in 2 divided doses in 4 phases in a crossover manner with a washout period of 7 days in between.Primary efficacy measure was responder rate (50% decrease in mean migraine frequency/intensity). Secondary ef-ficacy measures included reduction in mean monthly frequency, intensity, duration, rescue medication use, migraineassociated symptoms, and adverse events.

Statistical analysis.—Analysis was on intention to treat basis. Data were analyzed as correlated data. Generalizedestimation equation was used to compute overall mean standard deviation and 95% confidence intervals for eachof the outcome variables. Bonferroni’s correction done for multiple comparisons. P value of <.017 was taken assignificant.

Results.—Fifty-seven patients comprised the intent-to-treat population. Four patients withdrew from the studyat various phases, none because of the side effects. Responder rate for frequency was significantly higher fortopiramate versus placebo (63% vs 30%, P < .001), and versus lamotrigine (63% vs 46 %, P = .02). For intensityof headache also a responder rate of topiramate versus placebo (50% vs 10%, P < .001), and versus lamotrigine(50% vs 41%, P = .01) was observed. Topiramate showed statistically significant benefits (P < .017) in most of thesecondary efficacy measures while lamotrigine was beneficial for reduction in headache frequency, and migraineassociated symptoms. Adverse events were similar.

Conclusion.—Low-dose topiramate is efficacious in migraine prophylaxis as compared to both placebo andlamotrigine. Lamotrigine in low doses might be beneficial for headache frequency; however, longer trials are requiredto establish its efficacy on the intensity and frequency of migraine.

Key words: low dose, topiramate, lamotrigine, migraine prophylaxis

(Headache 2007;47:402-412)

From the Department of Neurology, All India Institute of Med-ical Sciences, New Delhi, India.

Address all correspondence to Dr. Sumit Singh, Departmentof Neurology, 707 Neurosciences Centre, All India Institute ofMedical Sciences, Ansari Nagar, New Delhi, India.

Accepted for publication May 2, 2006.

Migraine is a common neurological disorder as-sociated with significant disability, adversely affectingactivities of daily activity, work-related productivity,and significant impairment of quality of life.1,2 It hasbeen reported that it affects about 17% of women, 6%of men, and 4% of children yearly.3 Migraine patientsmay not seek healthcare, or may not be properly di-agnosed. Among patients who are treated, less than

402

Headache 403

one-third report consistently effective results withtheir current pharmacological regimens.4 Abuse ofacute therapies can lead to a chronic headache disor-der and may be a source of severe gastrointestinal andsystemic side effects. Preventive strategies are moreimportant in the management of migraine.5,6 Pre-ventive medications by reducing headache frequency,need for acute treatment, and side effects thereof, maysignificantly improve the quality of life in migraine.

It has been hypothesized that migraine is a dis-order resulting from neuronal hyperexcitability.7 Thishas led to exploration of the new antiepileptic drugs inthe prophylaxis of this pathology.8 Recent studies sug-gest that topiramate may modulate trigeminovascularsystem, which has an important role in migraine patho-genesis.9 Topiramate has multiple mechanisms of ac-tion, including state-dependent blockade of sodiumchannels, enhancement of GABAA receptor-mediatedinhibition, antagonism of glutamate at [α]-amino-3-hydroxy-5-methylisoxazole-4-propionic acid/kainatereceptors, and inhibition of high voltage-activated(l-type) calcium channels.10-12 These actions, individ-ually or in combination, may contribute to migraineprevention. Studies have shown role of high dose(100-200 mg/day) topiramate in migraine prophy-laxis.13,14 However, at these doses there are significantside effects.8

Lamotrigine blocks voltage-sensitive sodiumchannels, leading to inhibition of neuronal releaseof glutamate. Release of glutamate may be essentialin the propagation of spreading cortical depression,which some believe is central to the genesis of migraineattacks.15 Despite some open label studies showingbenefit in migraine with lamotrigine, a recent random-ized trial failed to replicate these benefits.15-17

In order to address the efficacy of low dose topira-mate in migraine and assess its efficacy as a migraineprophylactic and compare this with that of lamotriginein Indian population, we carried out the present ran-domized, double blind placebo controlled crossovertrial.

MATERIAL AND METHODSThe present study was undertaken at the headache

clinic of a tertiary referral center in India from June2003 to April 2004. The recruitment period was from

June 2003 to December 2003 and last follow-up wascompleted in April 2004. Consecutive patients pre-senting with complaint of headache and diagnosedas having migraine (with or without aura) accordingto International Headache Society criteria18 were re-ferred to one of the investigators. The diagnosis wasreconfirmed and a predesigned migraine performa in-cluding demographic and clinical details of the pa-tients was filled. The patients were recruited based onthe following inclusion and exclusion criteria after ob-taining an informed written consent, conforming to thecurrent revision of the declaration of Helsinki. The in-stitutional ethics committee approved the trial.

Inclusion Criteria.—

1. Diagnosis of migraine (with or without aura) ac-cording to International Headache Society cri-teria.

2. Duration of disease: at least 1 year.3. Frequency of attacks: 4 or more migraine

headache attacks per month and less than orequal to 10 attacks per month.

4. Each attack separated by pain free interval ofat least 48 hours.

5. Age at onset less than 50 years.6. Age at entry 18 to 65 years.7. Females of childbearing age group that are nei-

ther pregnant nor lactating and are ready to usereliable methods of contraception during thestudy.

8. The concomitant migraine prophylactics with-drawn 1 month prior to entry into trial.

9. Patient being able to fill a headache diary suc-cessfully and reliably.

Exclusion Criteria.—

1) Experienced headaches other than migraine.2) Had migraine onset after the age of 50 years.3) Overused acute migraine treatments (>8 treat-

ment days per month of ergots, NSAIDs, ortriptans).

4) Used β-blockers, tricyclic antidepressants,anti-epileptic drugs, calcium channel block-ers, monoamine oxidase inhibitors, dailynonsteroidal anti-inflammatory drugs, high-dose magnesium supplements (600 mg/day),

404 March 2007

Patients included in the trial n = 60

Topiramate arm n=30

Lamotrigine arm n=30

Topiramate drug n = 15

Topiramate placebo n = 15

Lamotrigine drugN = 15

Lamotrigine placebo n = 15

Topiramate placebo

Lamotrigine drug

Lamotrigine placebo

Topiramate drug

Lamotrigine placebo

Lamotrigine drug

Lamotrigine placebo

Topiramate placebo

Topiramate drug

Lamotrigine drug

Topiramate drug

Topiramate placebo

Fig 1.—Schematic representation of the flow of patients in the various treatment arms.

high-dose riboflavin (100 mg/day), cortico-steroids, local anesthetics, botulinum toxin, orherbal preparations 1 month prior to the study.

5) On antipsychotics/antidepressant drugs for3 months.

6) Alcohol or other drug dependence.7) Patients with nephrolithiasis or those who par-

ticipated in a previous topiramate or lamotrig-ine study, used topiramate or lamotrigine for2 weeks or longer, or used an experimental drugwithin 30 days of screening.

Study Design.—The present study was a singlecenter randomized double blind placebo controlled,4 phase crossover trial using topiramate and lam-otrigine in prophylaxis of frequent migraine. Afterthe screening, the subjects entered a 28-day prospec-tive phase during which they were taught to fill theheadache diary. Their reliability to do so was assessedby evaluating 2 such diaries filled by them. The patientswere also taught to indicate the severity of their mi-graine headache on a 100 mm VAS scale. The patientswho filled the headache diary reliably and fulfilled theinclusion and exclusion criteria were entered into thedouble blind phase of the study.

Randomization and Blinding.—Eligible patientswere randomized by a computer-generated random-ization schedule to receive:

(1) Lamotrigine 25 mg BD for 1 month or matchingplacebo OR

(2) Topiramate 25 mg BD for 1 month or matchingplacebo

The patients took the treatment assigned to themfor 1 month and then after a wash out period of 7 daysthe patients crossed over as follows: Active agent fol-lowed by matched placebo followed by the second ac-tive agent followed by the second placebo OR placebofollowed by matched drug followed by second placeboto be followed by the second drug (Figure 1).

Patients and clinicians were blinded to study med-ication with preprinted medication code labels. Sealedenvelopes containing the code labels with a tear offlabel concealing the randomization number were pro-vided to the investigator. Placebo was identical inappearance and packaging to active drug. As the lam-otrigine and topiramate tablets were different in ap-pearance 2 different placebos were used. Study medi-cation was packaged and labeled according to a medi-cation code schedule generated before the trial. Eachpackage had all 4 medications numbered according tothe phase of the trial. Each bottle had a 2-part tear-offlabel; study medication identification was concealedand could be revealed only in case of emergency.The treatment assignments were not revealed to studypatients, investigators, and the clinical staff until all

Headache 405

patients had completed therapy and the database hadbeen finalized.

The patients were allowed to take rescue medica-tions (diclofenac potassium and paracetamol combina-tion tablets) at their choice supplied to them in pack-ages of 24 each. The frequency, severity, and symptomsof all headaches or auras were recorded by each pa-tient in a diary, which was then transcribed into thepatient’s case record form at each clinical visit, whichtook place at the end of each month. Across all treat-ment groups, the patients themselves judged the typeof headache they had during the treatment course (mi-graine or nonmigraine headache), and transcribed itin the headache diaries. Each patient recorded theamount of rescue medication used, the severity ofheadache on the visual analogue scale, and the sideeffects.

Hematological investigations, serum chemistry(including glucose, sodium, potassium, serum urea ni-trogen, and total protein levels), liver function tests,and urinalysis were performed for all patients at thebeginning of each phase and physical examinationswere performed at the beginning and end of the study.Hematological and serum chemistry tests were per-formed at the first, third, and final visits of the double-blind phase. Liver function tests were performed eachclinic visit. Vital signs and weight were recorded ateach clinic visit. The adverse events during the studywere recorded routinely on the headache diary; the pa-tient was assessed at the onset of event and followedup until resolved or until a clinically stable endpointwas achieved.

Efficacy Measures.—The headache diary was usedfor outcome assessment. One of the investigators, whowas blinded, collected the headache diary at the endof every 4 weeks of a phase of the study. The pri-mary efficacy measures were a comparison amongthe topiramate and matching placebo; lamotrigineand matching placebo; and lamotrigine and topira-mate groups in terms of responder rate for frequencyof migraine headache attacks per month and meanheadache intensity for the migraine headaches indi-vidually (more than 50% reduction in baseline mi-graine headache frequency or intensity). The inten-sity of the migraine headache was defined and overallassessment for the entire duration of the headache.

Secondary efficacy endpoints included the reductionin migraine attack frequency, mean/median durationof migraine headaches, occurrence of migraine asso-ciated symptoms (photophobia, phonophobia, nau-sea, and vomiting), mean number of associated symp-toms per headache, response to rescue medication interms of intensity, and the total units of rescue medi-cation used. All efficacy measures were prospectivelydesignated.

Safety Evaluations.—Safety was assessed by reportsof adverse events communicated historically duringvisits, as transcribed on headache diaries, physical andneurological examination, and clinical laboratory testsas outlined earlier.

Statistical Analysis.—The demographic and clinicaldata were recorded on a predesigned performa andmanaged on a Microsoft Excel worksheet. All entrieswere crosschecked for errors. The analysis was doneon intention to treat basis and all patients who made anefficacy assessment were included in the analysis. Sincethe study design was crossover, in which each subjectreceived all treatments in a specified order, the groupswere not considered independent and the data wereanalyzed as correlated data. We used repeated mea-sures analysis using generalized estimation equationapproach to compute overall mean standard deviationand 95% confidence intervals for each of the outcomevariables (quantitative or binary). We used the samemethod for comparison between the 4 groups and cor-rected the “P” values using Bonferroni correction formultiple comparisons. As we used 3 comparisons be-tween the groups a P value of <.017 was taken as sig-nificant. STATA 8.0 statistical software was used fordata analysis. To detect a 34% difference in respon-der rate for frequency between topiramate phase andplacebo phase a sample size of 41 patients would havebeen sufficient. With the present sample size of 60 thestudy had 99% power to detect a significant differencebetween topiramate and placebo.

RESULTSWe screened 129 patients (97 females and 32

males) of migraine in the trial period. The first 60patients (47 females and 13 males) who fulfilled theexclusion and inclusion criteria and were able toreliably transcribe baseline diary assessments were

406 March 2007

Randomized n= 60

Screened N=129

Exclusion (69): lack of contraception (32), low level of literacy (20), frequency > 10 or < 4 (14), other headaches (12), contraindicated drugs (10), and refusal to give informed consent

Topiramate phase Allocated (n=60) Received treatment (n=57), 3 patients withdrew before this phase adverse events (2), lack ofefficacy (1)

Intension to treat population n= 56 as withdrawing patient did not make efficacy assessment

Topiramate placebo phase Allocated (n=60) Received treatment (n=57), 3 patients withdrew before this phase due to adverse events

Intension to treat population n=57 as withdrawing patient made efficacy assessment

Lamotrigine phaseAllocated (n=60) Received treatment (n=57), 3 patients withdrew before this phase adverse events (2), lack of efficacy (1)

Lamotrigine placebo phase Allocated (n=60) Received treatment (n=57), 3 patients withdrew before this phase due to adverse events (2), lack of efficacy (1)

Intension to treat population n=57 as withdrawing patient made efficacy assessment

Intension to treat population n= 56 as withdrawing patient did not make efficacy assessment

Follow up n=57 1 patient withdrew adverse event (pain in lower limbs)

Follow up n=57 1 patient withdrew adverse event (sedation)

Follow up n=57 1 patient withdrew, lack of efficacy

Follow up n=57 1 patient withdrew adverse event (sedation)

Fig 2.—Flowchart of randomized patients (consort statement).

randomized. As it was a crossover design all pa-tients received all interventions. The reasons for ex-clusion are mentioned in Figure 2. The mean (range)(SD) age of randomized patients was 29.41 (16-48) (±7.71) years with duration of symptoms for5.08 (1-15) (±4.44) years and a mean headache fre-quency of 6.98 (4-10) (±2.19) episodes per month.The intention to treat population had at least 1postbaseline efficacy assessment. The baseline char-acteristics of the randomized patients are given

in Table 1. At the time of analysis it was foundthat the patient groups were age and sex matched.During the 20-week period 4 patients withdrew fromthe trial. Three patients (5%) withdrew due to side ef-fects to the study medication (Table 2) and 1 patientin placebo group due to lack of efficacy. Four patients(6.7%) did not give efficacy assessment each in top-iramate and placebo matching to lamotrigine phasesand 3 patients (5%) did not make efficacy assessmentsfor lamotrigine and placebo topiramate phases. So the

Headache 407

Table 1.—Headache Characteristics of Randomized Patients

Headache Characteristics Number of Cases Percentage

Sex (female) 47 78.33Aura 19 31.67Premigrainous symptoms 7 11.17Premigrainous depression 18 30Improvement with tea 3 5.00Improvement with analgesics 34 57Improvement with sleep 19 31.67

intention to treat population was (n = 56) in topira-mate and placebo matching to lamotrigine phases and(n = 57) in lamotrigine and placebo matching to topi-ramate phases.

Efficacy Measures.—The primary outcome mea-sure, the responder rate for frequency (≥50% reduc-tion in monthly migraine frequency) for the topira-mate group versus matching placebo was 63% versus30% (P < .001 95% CI 0.18-0..46); for lamotrigineversus placebo 46% versus 34% (P < .093, 95% CI0.02-0.26), and for topiramate versus lamotrigine 63%versus 46% versus 34% (P < .019, 95% CI 0.03-0.31).The responder rate for headache intensity (≥50% re-duction in mean migraine intensity) for the topiramategroup versus matching placebo was 50% versus 10%

Table 2.—Results: Primary and Secondary Efficacy Measures (Patient Number (N) in Each Phase With Mean/Percentage andStandard Deviation)

Placebo Lamotrigine Lamotrigine Topiramate Placebo Topiramate

Efficacy Measure N Mean/% SD N Mean/% SD N Mean/% SD N Mean SD

PrimaryResponders (mean frequency) 56 34 0.47 57 0.46 0.50 56 0.63 0.49 57 0.30 0.46Responders (mean intensity on visual

analogue scale (VAS))49 0.14 0.35 47 0.21 0.41 36 0.50 0.51 50 0.10 0.30

SecondaryReduction in mean frequency/month 56 2.23 2.72 57 3.40 2.70 56 4.21 2.63 57 2.16 2.71Mean duration (hours) 56 11.29 9.3 57 12.99 13.27 56 6.75 1.18 57 12.56 10.67Intensity on VAS 49 67.03 17.6 47 61.21 18.39 36 54.66 18.53 50 67.99 17.29Phonophobia episodes/month 50 3.84 2.99 46 3.09 2.50 36 2.83 2.01 50 3.80 2.41Photophobia episodes/month 50 3.64 2.92 46 2.91 2.40 36 2.94 2.10 50 3.88 2.40Rescue medication use (number/month) 60 3.95 4.12 60 2.75 3.39 60 1.80 2.59 60 4.65 4.79Improvement in intensity with rescue

medication45 69.81 18.52 40 7.70 16.49 28 75.53 15.96 47 69.25 19.88

Number of aura/month 60 0.05 0.29 60 0.02 0.13 60 0.02 0.13 60 0.12 0.45Drug side effects 60 0.07 0.25 60 0.10 0.30 60 0.17 0.49 60 0.12 0.37

(P < .001, 95% CI 0.17-0.46); for lamotrigine versusplacebo 21% versus 14% (P < .093, 95% CI 0.02-0.26),and for topiramate versus lamotrigine 63% versus 46%versus 34% (P < .019, 95% CI 0.03−0.31).

Secondary endpoints also demonstrated statisti-cally significant changes. Among the secondary effi-cacy measures both topiramate and lamotrigine wereassociated with a significantly greater decrease thanmatching placebos in mean (SD); monthly migrainefrequency: topiramate versus placebo 4.21 (±2.63)versus 2.16 (±2.71) (P < .001, 95% CI [(−2.69) to(−1.27)]), lamotrigine versus placebo 3.40 (±2.70) ver-sus 2.23 (±2.72) (P .002, 95% CI [(−1.84) to (−0.42)].The difference in terms of mean decrease in migrainefrequency was significant for topiramate versus lamot-rigine P value .019 but did not remain significant afterthe Bonferroni correction (Table 3).

Topiramate was associated with statistically signif-icant reductions in mean duration of headache com-pared both with placebo and lamotrigine; mean (SD)topiramate versus placebo 6.75 (±10.18) hours versus12.56 (±10.67) hours (P value .001, 95% CI 2.49-9.21);topiramate versus lamotrigine 6.75 (±10.18) hours ver-sus 12.99 (±13.27) hours (P < .001, 95% CI 2.86-9.58). The difference was nonsignificant for lamotrig-ine versus placebo (P .326). The number of severeheadaches as subjectively recorded by the patient was

408 March 2007

Table 3.—Results: Comparison Between Various Phases Showing Regression Coefficient, Significance, and 95% ConfidenceInterval

Lamotrigine Versus Placebo Topiramate Versus Placebo Topiramate Versus lamotrigine

Efficacy Regression Coefficient Regression Coefficient Regression CoefficientMeasure (95% CI) P value (95% CI) P value (95% CI) P value

PrimaryResponders (mean

frequency)−0.12 ([0.26]-[−0.02]) .093 −0.32 ([−0.46]-[−0.18]) <.001 −0.17 ([−0.03]-[−0.31]) .02

Responders (meanintensity on VAS)

−0.07 ([−0.22]-[0.08]) .340 −0.40 ([−0.57]-[0.025]) <.001 −0.30 ([0.46]-[−0.17]) <.001

SecondaryReduction in mean

frequency/month−1.13 ([−1.84]-[−0.42]) .002 −1.98 ([−2.69]-[−1.27]) <.001 −0.85 ([−0.14]-[−1.56]) .019

Mean duration (hours) −1.68 ([−5.04]-[1.67]) .326 5.85 (2.49-9.21) .001 6.22 (2.86-9.58) <.001Intensity on VAS 5.19 (−0.14-10.40) .056 13.08 (7.33-18.83) <.001 7.09 (1.31-12.87) .016Phonophobia

episodes/month0.84 (0.20-1.48) .010 1.63 (0.92-2.33) <.001 0.84 (0.13-1.55) .020

Photophobiaepisodes/month

0.78 (0.12–1.44) .020 1.48 [(0.76)-(−20)] <.001 0.46 (−0.26-1.19) .209

Total number ofsymptoms/month

2.82 (0.65)–5.00) .011 5.49 (3.09-7.89) <.001 2.46 (0.05-4.87) .046

Mean number ofsymptoms/headache

0.59 (−0.10)-(1.28) .092 0.47 (−0.28)-(1.22) .217 1.61 (0.60)-(4.36) .347

Rescue medication use(number/month)

1.16 (0.20)-(2.15) .018 2.59 (1.60)-(3.58) <.001 0.99 (−0.01)-(1.99) .053

Improvement inintensity with rescuemedication

−2.45 (−7.88)-(−2.97) .375 −9.20 (−15.35)-(−3.06) .003 −5.96 (−12.22)-(0.30) .062

Drug side effects 11.01 (2.09)-(19.93) .017 8.56 (−0.33)-(17.45) .059 −6.0 (−15.43)-(3.43) .212

also significantly less in the topiramate group: topira-mate versus placebo 2.47 (±1.54) versus 2.80 (±1.65)(P value .005, 95% CI 0.25-1.38). The difference formild and moderate headaches and all headaches inother groups was not significant.

The mean (SD) headache intensity on VAS wassignificantly reduced for treatment with topiramateversus placebo, 54.66 (±18.53) versus 67.99 (±17.29)(P < .001, 95% CI 7.33-8.83) and topiramate versuslamotrigine 54.66 (±18.53) versus 61.21 (±18.39) (P.016, 95% CI 1.31-12.87). The difference was not sig-nificant for lamotrigine in comparison to placebo. Top-iramate treated patients had a reduction in the mean(SD) doses of the rescue medications 1.80 (±2.59) ver-sus 4.65 (±4.79) (P < .001, 95% CI 1.60-3.58) andimprovement in the mean response to medication onVAS 75.53 (±15.96) versus 69.25 (±19.88) (P value.003) as compared to placebo. The total rescue med-ication use difference in lamotrigine versus placebophases was significant (P .018) but the significance was

lost after applying the Bonferroni correction. Topira-mate decreased the total occurrence of migraine re-lated photophobia, phonophobia, and nausea as com-pared to placebo (P < .001) while lamotrigine reducedonly photophobia as compared to placebo (P = .010).The reduction in aura was significant only for topira-mate versus placebo (P = .031) but the significancedisappeared after applying the Bonferroni correction.We also did a sensitivity analysis for the primary end-points assuming that all the patients who dropped outwere nonresponders and the difference still remainedsignificant for topiramate versus placebo both in termsof intensity (P < .001) and frequency (P < .001). Thedifference was still significant for topiramate versuslamotrigine for headache intensity (P < .001) and thesignificance for frequency (P .037) disappeared afterapplying the Bonferroni correction.

Safety Measures.—Treatment-related side effectsoccurred in 10% patients treated with lamotrigine andincluded rash, giddiness, sleepiness, gastrointestinal

Headache 409

Table 4.—Adverse Effects in Number in Each Phase With Number of Patients Discontinuing Due to Side Effect in Parentheses

Adverse Event Lamotrigine (95) Placebo Lamotrigine Topiramate Placebo Topiramate

Sleepiness and concentration difficulty 2 (1) 0 3 (1) 0Paresthesias 2 3 1Gastrointestinal intolerance 2 1 3 2Anorexia 1 2 1 2Giddiness 2 1 2 1Rash 2 1 0 0Palpitations 0 2 0 0Menorrhagia 0 0 0 1Hair loss 0 0 0 1Pain in lower limbs 0 1 (1) 0 0

intolerance and in 7% patients with matching placebo,with anorexia and palpitation being most frequent. Inthe topiramate group, 15% patients had side effects ofwhich most common were paresthesia, sleepiness, andgastrointestinal intolerance occurring in 5.0% patientswhile on matching placebo 10% patients had side ef-fects most commonly gastrointestinal intolerance andanorexia (3.3%) patients each. The difference in oc-currence of side effects was not significant in topira-mate versus matching placebo (P = .059), lamotrigineand matching placebo (P = .17) and lamotrigine versustopiramate (P = .212) (Table 4).

DISCUSSIONIn the present study, low dose topiramate (50

mg/day) showed significant efficacy as a prophylac-tic agent for migraine as compared to placebo andlow dose lamotrigine (50 mg/day). Topiramate also re-duced headache intensity by 50% in as many as 40%more patients as compared to placebo. This gives aNNT of 3 (2-5) for 50% reduction in headache fre-quency for topiramate, which is comparable with theNNT of all anticonvulsants together (3.8).8

Reports from previously conducted trials of top-iramate in migraine have shown its statistically sig-nificant benefit as a prophylactic agent for migraine(Table 5).13,14,19-23 All these studies are parallel grouptrials where the results may vary with headache char-acteristics of patients as the groups have different pa-tients. In a crossover trial the patient population ex-posed to 2 interventions is the same and not merely

similar. To the best of our knowledge the present studyis the only trial using a crossover study design. The2 large trials, which evaluated 50, 100, and 200 mgdoses, failed to show significant benefits for 50 mg doseof topiramate with respect to primary endpoints;13,14

however, one trial using low-dose topiramate showedhighly significant results in chronic migraine.23 An-other open label study of 117 patients, with chronic andepisodic migraine, showed up to 77% responder rateswith low dose topiramate.24 Our study highlights theefficacy of low dose topiramate in frequent migrainealso (frequency >4 per month).

A decrease in mean frequency of migraine/monthhas been used as primary outcome measure for effi-cacy in most of the previous studies with benefits inthe range of reduction in migraine frequency by 2-2.5headaches/month.13,14,19-23 This may not be a clinicallysubstantial benefit in patients with more frequent mi-graine. Responder rate is clinically a more robust mea-sure of efficacy of a drug as a migraine prophylactic.Ours is the first study to use responder rate as the pri-mary endpoint and show significant efficacy in termsof responder rates for both frequency and intensity.We showed significantly high responder rates as com-pared to previous studies with 50-mg topiramate. Inour study, the responder rates, even in both placebophases, are 8-10% higher than reported in other recentstudies. This may be due to a more methodological as-sessment and management of migraine in a HeadacheClinic with stress on patient education, awareness oftrigger factors as compared to routine managementby general practitioners. A lack of prior exposure to

410 March 2007

Table 5.—Randomized Controlled Trials of Topiramate in Migraine Prophylaxis

Study Duration Number of Dose Results (PrimaryAuthor (95) Patients Used Endpoint) Remarks

Silberstein et al2004. (13)

26 weeks 483 50 mg, 100 mg, 200 mg Mean frequency red. −2.1(100 mg), −2.4 (200mg) P <.001

50 mg no significantresponse, parallelgroup

Brandes et al2004. (14)

26 weeks 487 50 mg, 100 mg, 200 mg Mean frequency red. −2.1(100 mg), −2.3 (200mg) P <.001

50 mg no significantresponse, parallelgroup

Mei 2004. (19) 16 weeks 115 100 mg Mean frequency red.−2.6, significant

50 mg dose not tried,parallel group

Deiner et al 2004.(20)

575 100 mg, 200 mg,propanolol 160 mg

Mean frequencysignificant reduction

Topiramate 100 mgsimilar to propanolol160 mg

Edwards et al 2003.(21)

20 weeks 70 Target dose 200 mg Frequency 3.2 versus 3.8,P.001

50 mg dose not tried,parallel group

Storey et al 2001.(22)

16 weeks 40 Target dose 200 mg Frequency 3.3 versus 3.8,P .002

50 mg dose not tried,parallel group

Silverstini et al2002. (23)

8 weeks 28 50 mg Headache days t versus P8.1 versus 20.6, P .0007

Shows effectiveness oflow-dose topiramate

migraine prophylaxis may have generally improvedthe responder rates across all treatment groups. Mi-graine associated symptoms are an important cause ofmigraine related morbidity. Most of the previous trialshave not used a decrease in migraine-associated symp-toms as a secondary endpoint. Our study demonstratesa significant reduction in migraine associated nausea,photophobia, and phonophobia. We have also shownthat topiramate enhances the efficacy of rescue medi-cations. Similar results have been reported in post hocanalysis of 2 studies with topiramate 100 mg.25 Thewithdrawal rate with low dose topiramate was verylow (1.8%), indicating excellent tolerability. The mainside effects, paresthesias, sedation, and gastrointesti-nal intolerance were similar to previous studies.13,14

In the lamotrigine phase, the responder rate bothin terms of headache frequency and intensity were notsignificant as compared to matching placebo. Open la-beled studies16 have shown efficacy of lamotrigine inreducing migraine frequency, which has not been re-produced in randomized trials.15 In our study, lamo-trigine significantly decreased mean monthly fre-quency as compared to placebo but the responder ratein lamotrigine group versus the placebo group was notstatistically significant. This demonstrates the impor-tance of using responder rate as a primary outcome

measure results in highlighting of clinically substan-tial benefits. Alternatively, it is also possible that sam-ple size was small to detect a difference in responderrate in lamotrigine group as compared to placebo. Wealso showed that lamotrigine significantly improvedmigraine related photophobia and giddiness as com-pared to placebo. This is consistent with results of aprevious observational study, which showed reductionin migraine frequency by 50% and migraine relatedvertigo by 66%.26 The reduction in migraine relatedaura was not significant with lamotrigine in our studyas compared to some open labeled studies, which hadshown significant reductions up to 80%.17,27 Becausein our study only 19 (31%) patients had migraine withaura, we may not have been able to detect these dif-ferences due to small numbers. It is possible that lam-otrigine may have a beneficial action on selective mi-graine related phenomena, as partly demonstrated inour study.

This is the first randomized trial to compare top-iramate with lamotrigine. We showed that topira-mate in comparison with lamotrigine caused signifi-cant responder rates in terms of headache intensityand frequency. However, the significance of responsein terms of frequency disappeared after Bonferronicorrection, suggesting that it may result from multiple

Headache 411

comparisons. Among the secondary efficacy measurestopiramate was significantly better than lamotrigine inreduction of mean duration of headache and frequencyof headaches.

The small numbers of patients recruited in ourstudy and short duration of follow up are the rela-tive limitations of our study. The number of patientsdespite being apparently small had adequate power(99%) to show significant differences between topi-ramate and placebo. In most previous studies the re-sponse to topiramate has become apparent within thefirst month of treatment; therefore, the short durationof follow-up may not be a significant limitation forassessment of efficacy of topiramate.13,14 The presentstudy underscores the efficacy of low doses of topira-mate, which is likely to be associated with less severeside effects. The importance of studies with longer du-ration of treatment, however, cannot be overempha-sized. On the basis of the present study the efficacy oflamotrigine cannot be refuted as a prophylactic for mi-graine, as the study duration was short and the powerof the study was not sufficient to make this conclusion.

COMMENTSOur study demonstrates the efficacy of low-dose

topiramate in episodic migraine. This may be the rea-son to initiate further trials with low-dose topiramate,in migraine prophylaxis, thereby reducing the concernfor potential side effects, a limiting factor for the useof topiramate. We have also demonstrated that lamot-rigine is not beneficial in migraine prophylaxis, whichis inconsistent with previous studies. However, beforerelegating this drug to the background, further studieswith adequate numbers and duration using lamotrig-ine in migraine prophylaxis are warranted.

The present work was presented in the MigraineTrust International Symposium, London, UK, and theabstract will be published in the forthcoming issue ofCephalalgia.

Conflict of Interest: None

REFERENCES

1. Terwindt GM, Ferrari MD, Tijhuis M, Groenen SM,Picavet HS, Launer LJ. The impact of migraine on

quality of life in the general population: The GEMstudy. Neurology. 2000;55:624-629.

2. Lipton RB, Hamelsky SW, Kolodner KB, Steiner TJ,Stewart WF. Migraine, quality of life, and depression:A population-based case-control study. Neurology.2000;55:629-635.

3. Silberstein SD, Lipton RB, Goadsby PJ. Headachein Clinical Practice. 2nd ed. Oxford, England, MartinDunitz; 2002.

4. Brandes JL. Global trends in migraine care: Resultsfrom the MAZE survey. CNS Drugs. 2002;16(suppl1):13-18.

5. Silberstein SD. Practice parameter: Evidence-basedguidelines for migraine headache (an evidence-basedreview): Report of the Quality Standards Subcommit-tee of the American Academy of Neurology. Neurol-ogy. 2000;55:754-762.

6. Limmroth V, Katsarava Z, Fritsche G, Przywara S,Diener HC. Features of medication overuse headachefollowing overuse of different acute headache drugs.Neurology. 2002;59:1011-1014.

7. Welch KM, D’Andrea G, Tepley N, et al. The conceptof migraine as a state of central neuronal hyperex-citability. Neurol Clin. 1990;8:817-828.

8. Chronicle E, Mulleners W. Anticonvulsant drugsfor migraine prophylaxis. Cochrane DatabaseSyst Rev 2004, Art. No.: CD003226. DOI:1.1002/14651858.CD003226.pub2.

9. Storer RJ, Goadsby PJ. Topiramate inhibits trigemi-novascular traffic in the cat: A possible locus ofaction in the prevention of migraine. Neurology.2003;60(suppl 1):A238-A239.

10. White HS. Clinical significance of animal seizuremodels and mechanism of action studies of potentialantiepileptic drugs. Epilepsia. 1997;38(suppl 1):S9-S17.

11. White HS, Brown SD, Woodhead JH, Skeen GA,Wolf HH. Topiramate enhances GABA-mediatedchloride flux and GABA-evoked chloride currents inmurine brain neurons and increases seizure thresh-old. Epilepsy Res 1997;28:167-179.

12. Shank RP, Gardocki JF, Streeter AJ, Maryanoff BE.An overview of the preclinical aspects of topiramate:Pharmacology, pharmacokinetics, and mechanism ofaction. Epilepsia. 2000;41(suppl 1):S3-S9.

13. Silberstein SD, Neto W, Schmitt J, Jacobs D; MIGR-001 Study Group. Topiramate in migraine preven-tion: Results of a large controlled trial. Arch Neurol.2004;61:490-495.

412 March 2007

14. Brandes JL, Saper JR, Diamond M, et al. MIGR-002Study Group. Topiramate for migraine prevention:A randomized controlled trial. JAMA. 2004;291:965-973.

15. Steiner TJ, Findley LJ, Yuen AW. Lamotrigine ver-sus placebo in the prophylaxis of migraine with andwithout aura. Cephalalgia. 1997;17:101-102.

16. D’Andrea G, Granella F, Cadaldini M, Manzoni GC.Effectiveness of lamotrigine in the prophylaxis of mi-graine with aura: An open pilot study. Cephalalgia.1999;19:64-66.

17. Lampl C, Buzath A, Klinger D, Neumann K. Lam-otrigine in the prophylactic treatment of migraineaura—a pilot study. Cephalalgia. 1999;19:58-63.

18. Headache Classification Committee of the In-ternational Headache Society. Classification anddiagnostic criteria for headache disorders, cranialneuralgia, and facial pain. Cephalalgia. 1988;8(suppl7):1-96.

19. Mei D, Capuano A, Vollono C, et al. Topiramatein migraine prophylaxis: A randomised double-blindversus placebo study. Neurol Sci. 2004;25:245-250.

20. Diener HC, Tfelt-Hansen P, Dahlof C, et al.MIGR-003 Study Group. Topiramate in migraineprophylaxis—results from a placebo-controlled trialwith propranolol as an active control. J Neurol.2004;251:943-950.

21. Edwards KR, Potter DL, Wu SC, Kamin M, Huli-han J. Topiramate in the preventive treatment ofepisodic migraine: A combined analysis from pilot,double-blind, placebo-controlled trials. CNS Spectr.2003;8:428-432.

22. Storey JR, Calder CS, Hart DE, Potter DL. Top-iramate in migraine prevention: A double-blind,placebo-controlled study. Headache. 2001;41:968–975.

23. Silvestrini M, Bartolini M, Coccia M, Baruffaldi R,Taffi R, Provinciali L. Topiramate in the treatment ofchronic migraine. Cephalalgia. 2003;23:820-824.

24. Kowacs PA, Piovesan EL, Watzko DR, FameliH, Werneck LC. Low-dose topiramate in migraineand transformed migraine prophylaxis. Cephalalgia.2003;23:721 [Abstract].

25. Dodick D, Hulihan J, Wu S. Do migraine preven-tive medications enhance the efficacy of triptans?Analyses from topiramate migraine trials. Headache.2005;45:820. [Abstract].

26. Bisdorff AR. Treatment of migraine related ver-tigo with lamotrigine an observational study. BullSoc Sci Med Grand Duche Luxemb. 2004;2:103-108.

27. Pascual J, Caminero AB, Mateos V, et al. Preventingdisturbing migraine aura with lamotrigine: An openstudy. Headache. 2004;44:1024-1028.