Topiramate (Topamax) for Migraine Prophylaxis

Reduces frequency of headaches in episodic migraine

Benefits in NNT

1 in 4 were helped (>50% reduction in frequency of headaches)
24% higher rate of responders (proportion of subjects who reported >50% reduction in frequency of headaches) compared to placebo
1.2 fewer migraine attacks per month

Harms in NNT

Not adequately reported
Not adequately reported
View As:

Efficacy Endpoints

Frequency of episodic migraines in a 4-week period; response to prophylaxis (subjects reporting ≥50% reduction in headache frequency)

Harm Endpoints

Anorexia, paresthesia, fatigue, nausea, weight loss, taste disturbances


The global prevalence of migraine headaches is estimated to be about 15%, consisting mostly of young and middle-aged adults.1 Migraine headaches not only cause a burden of pain and suffering, but also result in significant loss of productivity. For example, in the European Union migraines are estimated to cost nearly 50 billion Euros in lost productivity in terms of aggregate direct and indirect costs to society annually.2 A reliable method of preventing symptoms would be beneficial in terms of quality of life as well as societal productivity.3

There are several options for medications that patients can take to mitigate the effects of migraines. One medication that has been recommended for migraine prophylaxis is topiramate.4 Topiramate is classified as an antiepileptic, and while the exact mechanism by which antiepileptics affect migraine headaches is unclear, it has been suggested that topiramate prevents migraine attacks by modulating a variety of neurotransmitters.5

Here we will discuss data from a Cochrane systematic review examining the efficacy of topiramate as a pharmaceutical first line agent for migraine prophylaxis. There are many reasons for a patient to prefer one agent over another, such as desire for pregnancy, comorbid conditions, and side effect tolerance. Therefore, we chose to report the data on the subset of patients in whom topiramate was compared to placebo rather than against alternate agents.

The Cochrane systematic review and meta-analysis discussed here included 17 randomized controlled trials involving a total of 2282 adult patients (>16 years old) who suffered from episodic migraines, which are defined to be discrete flares of migraine symptoms.6 The included studies were required to be prospective, controlled trials of self-administered topiramate taken on a regular basis for preventing migraine attacks. The study endpoints included headache frequency, the proportion of responders (subjects with ≥50% reduction in headache frequency), and adverse events.

Treatment with topiramate increased the number of responders (odds ratio: 3.18, 95% CI, 2.1 to 4.82; absolute risk difference: 24%; number needed to treat [NNT]: 4, 95% CI 3 to 6). The included studies used topiramate in doses of 50 mg, 100 mg, or 200 mg per day. The meta-analysis found that topiramate significantly reduced the frequency of headaches in the treatment group within 28 days, as compared to placebo (mean difference for headache frequency: -1.20, 95% confidence interval (CI) -1.59 to -0.80; 9 trials, 1737 subjects).

The results remained statistically significant when data were stratified by dose.5 Data from three studies that directly compared 50mg, 100 mg, and 200mg doses of topiramate favored the 100 mg. 100 mg resulted in an increased proportion of responders compared to the 50mg dose. Meanwhile, there was no significant superiority of the 200mg dose over the 100 mg dose.2, 5

Harms: The method of reporting adverse events varied from trial to trial and not all trials reported adverse events.

Only 3 trials included in the systematic review explicitly reported adverse events.6 It is thus likely that adverse events were underreported. The lack of adequate information about harms should be discussed with patients as part of the shared decision-making process when this medication is being considered for prophylaxis.

In the studies where adverse events were adequately reported, the risk of adverse events was higher in the topiramate group when compared to placebo. The adverse events were dose-dependent and more common in patients on the 200 mg per day regimen than those on the 100 mg per day regimen.6 The adverse events included anorexia (number needed to harm [NNH]: 12-17), fatigue (NNH: 12-25), memory problems (NNH: 12-25), nausea (NNH: 17), paresthesia (NNH: 2-3), taste disturbance (NNH: 7-14), and weight loss (NNH: 11-25).6 In the group of patients who took less than 100 mg topiramate per day, the incidence of most adverse effects was not statistically different compared to placebo, with the exception of weight loss and taste disturbance.

Another systematic review published in 2015 assessed the comparative effectiveness of drugs used for prophylaxis of migraine headaches also reported an increased risk of nausea and paresthesia with daily use of topiramate compared to placebo.6

Because of the inconsistencies in reporting the adverse events and the small sample size of the few trials that reported them, interpreting the harm numbers warrants caution. For this reason, we have not reported the NNH values in the summary table.


The most notable consideration before prescribing topiramate is that in addition to the patient, the patient's primary care physician and/or neurologist ought to be involved in decision-making. Topiramate is a maintenance medication for chronic use. The patient must be willing to take the medication daily, and their outpatient team must be willing to prescribe it. Coordinating care with a primary care physician or a neurologist will ensure that the treatment success and possible adverse events are monitored regularly, refills are ordered, etc. In addition, there are important considerations for women of childbearing age. There is an increased risk of cleft lip and cleft palate among infants exposed to topiramate during the first trimester (relative risk of 21.3, 95% CI:7.9 – 57.1), as compared to the risk in a background population of untreated women.7 It should be noted that palate formation often occurs before patients are even aware of early pregnancy.

While data from the Cochrane systematic review is largely in favor of Topiramate for migraine prophylaxis,6 biases in the individual studies, challenges in the clinical identification of migraine diagnosis, and underreporting and analysis of harms must be considered. FDA warnings regarding birth defects in association with topiramate use and drug costs are also discussed.

The review does not specifically rate the quality of evidence for each outcome. However, it states that the quality of the existing evidence is enough to allow a robust conclusion to be drawn regarding the efficacy of topiramate in preventing migraine attacks. Some of the included trials suffered from biases that could threaten the validity of the findings, of the 17 studies reviewed, nine were judged to have a high risk of bias for at least one item, including allocation concealment, blinding of participants, incomplete outcome data, and selective reporting. Some studies were also underpowered. Another threat to validity was the heterogeneity among the trials, originating from differences in study populations, treatment regimens, and study protocols. As noted in this systematic review, this resulted in increased uncertainty when evaluating the overall effect of topiramate in migraine prophylaxis.6, 7

Another factor that adds to subjectivity and bias is that the diagnosis of migraine headaches is largely clinical and based on a thorough physical examination and a careful medical history.8, 9 While some clinical criteria such as those proposed by the International Classification of Headache Disorders are frequently used, the inconsistencies and lack of agreement between clinicians and researchers point to the subjective quality of many of these criteria.8 Thus, in general, trials involving migraine headache tend to suffer from biases originating from imperfect case definition and differential misclassification of exposure. In addition, data from the clinical trials analyzed do not provide sufficient evidence for the use of topiramate in preventing other components of migraine attacks, such as aura symptoms, prodromal symptoms, or chronic migraines.

Summary: The 2012 guidelines from the American Academy of Neurology (AAN) and the American Headache Society supports the use of topiramate and recommends that it be offered for migraine prevention (Level A).4 The European Federation of Neurological Societies guideline has a similar recommendation.10

Considering data from this Cochrane review, a daily dose of 100mg is a reasonable target daily dose.6 Having a discussion of possible adverse events with patients prior to starting a prescription is encouraged. As noted above, a discussion with the patient’s primary care physician or neurologist is also prudent before starting the treatment.

Cost is another factor for many patients when considering a new therapy. As of November 2020, a one-month supply of topiramate (50 mg twice daily) costs $9.00-11.81.11, 12

Current dosing recommendations of topiramate for migraine prophylaxis approved by the FDA is to start at 25 mg a night for week 1, 25 mg twice a day for week 2, 25 mg in the morning and 50 mg in the evening on week 3, and finally 50 mg both morning and evening starting week 4.12 In patients with significant side effects, rapid discontinuation without tapering is favored.12

In summary, the existing evidence supports the prophylactic use of topiramate to reduce the frequency of headaches in episodic migraine. Although the risk of adverse events was not reported fully, the adverse events that were reported with topiramate use appear to be relatively minor. Therefore, we have rated this treatment Green (Benefits > Harms). This recommendation is consistent with existing practice guidelines from specialty societies. Future trials should focus on the safety of topiramate and better delineate the side effect associated with the use of this drug.

The original manuscript was published in Academic Emergency Medicine as part of the partnership between and AEM.


Bing Li, MD; Paniz Johari, MS; Carolina Camacho Ruiz, MD
Supervising Editors: Kabir Yadav, MD


April 5, 2021