General lack of use of placebo in prophylactic,randomised,controlled trials in adult migraine. A systematic review

Abstract

Background

The Clinical Trials Subcommittee of the International Headache Society (IHS) recommends that a placebo arm is included in comparative randomised clinical trials (RCTs) of multiple prophylactic drugs due to the highly variable placebo response in migraine prophylaxis studies. The use of placebo control in such trials has not been systematically assessed.

Methods

We performed a systematic review of all comparative RCTs of prophylactic drug treatment of migraine published in English from 2002 to 2014. PubMed was searched using the Cochrane Highly Sensitive Search Strategy for identifying reports of RCTs.

Results

A placebo arm was used in <10% (three of 31) of prophylactic RCTs in migraine. In only 7.1% (two of 28) of the comparative RCTs without placebo was one drug superior to another drug. Thus in 26 RCTs, including one study requiring more than 75,000 patient days, no difference was identified across treatment arms and conclusions regarding drug superiority could not be drawn.

Conclusions

The majority of comparative, prophylactic migraine RCTs do not include a placebo arm. Failure to include a placebo arm may result in failure to demonstrate efficacy of potentially effective migraine-prophylactic agents. In order to benefit current and future patients, the current strong tendency to omit placebo-controls in these RCTs should be replaced by adherence to the guidelines of the IHS.

Keywords

Prophylactic trials migraine placebo ethical implications

Introduction

Migraine is a chronic disorder with recurrent headache attacks. The global prevalence of migraine is estimated to be 15% (1,2). Many patients can be treated successfully with current acute, specific treatment, the triptans (5-HT1B/1D receptor agonists) (3). In a very large United States (US) epidemiological study (n = 126,576), it was found that 31% of patients with migraine have three or more migraine attacks per month. From this study, and expert opinion-based treatment guidelines, it was estimated that approximately 25% of all migraine sufferers should be offered preventive therapy (4,5). However, only 13% of all migraine patients received daily preventive drug treatment in the US study (4).

Current prophylactic drugs used in migraine, nonsteroidal anti-inflammatory drugs (NSAIDs) (6), antiepileptic drugs (7), β-adrenoceptor blockers (8), antiserotonin drugs (9), and calcium antagonists (10), were with the exceptions of pizotifen and methysergide (10), developed for other indications and then used, often by serendipity, and found somewhat effective in migraine. Subsequently, efficacy was confirmed in randomised, controlled trials (RCTs). Unfortunately, the efficacy of these drugs is relatively low while side effects are frequent. Migraine continues to represent a very large burden of disease worldwide (1,2), and there is a large and unmet need for new prophylactic drugs (11). Evidently, new drugs should be shown to be similar or superior to current drugs in terms of efficacy and tolerability.

Due to highly variable placebo responses in prophylactic migraine RCTs, ranging from 0% to 56% responder rates (12), the guidelines for controlled trials of drugs in migraine of the International Headache Society (IHS) (13) recommend that a placebo arm is included in comparative RCTs of multiple prophylactic drugs in order to determine if an apparent effect of one or more drugs is simply placebo related.

The aim of the present study was to systematically review recent comparative, prophylactic RCTs in migraine in order to assess the adherence to the IHS recommendation of placebo control.

Methods

The study was based on data from a previous review of migraine RCTs from 2002 to 2008 (14), which was further expanded to include studies published until 2014.

We searched PubMed for clinical trials of drug treatment for migraine published from 2002 to 2014.

The search was performed using the Cochrane Highly Sensitive Search Strategy for PubMed (as revised in 2008) (15) and with publication date limits ranging from 1 January 2002 until 31 December 2014, the search string thus being: ‘((‘2002/01/01’[Date - Publication]: ‘2014/12/31’[Date - Publication])) AND ((migraine AND (randomised controlled trial OR randomised OR placebo OR drug therapy OR randomly OR trial OR groups) NOT animals))”. The data search was carried out on 23 September 2015. Search, study selection and data extraction were performed by a single reviewer (AH).

Comparative RCTs (i.e. comparing the efficacy of two or more different drugs) of prophylactic treatment were specifically selected for the present review.

The following were excluded from the review: Studies on acute treatment of migraine, studies of combined pharmacological and non-pharmacological treatment, studies not reported in English, studies not concerning migraine exclusively, studies with non-clinical outcome measures only, studies of induced migraine, and studies of patients under the age of 18 years.

Results

The search retrieved a total of 6316 items. From these 31 comparative, prophylactic trials of acute treatment were excluded from further analysis in this review, while 26 prophylactic trials were included (16 –46). Placebo was used in only three of the prophylactic RCTs (9.7%).

In one of these two RCTs the two active drugs, topiramate 100 mg and propranolol 160 mg, were both superior to placebo and not different in efficacy (24). In the other RCT with low doses one active drug, topiramate 50 mg, was superior to the other drug, lamotrigine 50 mg, and placebo, whereas lamotrigine was not superior to placebo (28). In the third study, candesartan 16 mg and propranolol 160 mg were both superior to placebo but not different in efficacy (42).

In only two RCTs without placebo was one drug superior to another. One of these was a study comparing topiramate 50 mg to propranolol 80 mg (17), while the other was an atypical prophylactic study for the prevention of menstrually related migraine by treatment in the perimenstrual period, showing superior efficacy of frovatriptan compared to naproxen and transdermal oestrogen (27). In all the other 26 RCTs there was no difference between the new drug and the standard drug for the primary efficacy parameter, in most cases (21/26) migraine frequency. An overview of the included RCTs is shown in Table 1. A summary of the largest RCT (23) ever on migraine prophylaxis is shown in Table 2, illustrating comparability both for efficacy and tolerability among the three treatments.

Table 1.

Overview of the included studies.

Trial by first author, year (ref)	Size	D	Dur	Pla	Drug 1	Drug 2	Drug 3	Primary endpoint	Outcome
Ali, 2010 (16)	40	P	4	No	Topiramate 50 mg	Thioctic acid 300 mg	Combination	Reduction in migraine attack frequency per month	Significant change from baseline in combination group only
Ashtari, 2008 (17)	62	P	8	No	Topiramate 50 mg/day	Propranolol 80 mg/day	NA	Not specified	Topiramate>Propranolol for frequency, intensity and duration
Bartolini, 2005 (18)	49	P	4	No	Topiramate 25–75 mg/day	Valproate 250–750 mg/day	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Bavrasad, 2010 (19)	73	P	12	No	Topiramate 75 mg	Sodium valproate 200–600 mg	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Blumenfeld, 2008 (20)	59	P	12	No	Botox 100U	Divalproex 250 mg	NA	Reduction of headache days per month	Significant change from baseline, no difference between treatments
Bulut, 2004 (21)	76	C	12	No	Venlafaxin 37.5 mg/75 mg/150 mg	Amitriptyline 10 mg/25 mg/50 mg	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Cady, 2010 (22)	59	P	12	No	Topiramate 25–200 mg	Onabotulinumtoxin A (109 U avg)	NA	The treatment responder rate based on the Physician Global Assessment	Significant change from baseline, no difference between treatments
Cosentino, 2013 (43)	60	P	12	No	Sincronil (topiramate) 25 mg	Topamax (topiramate) 25 mg	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Diener, 2002 (23)	808	P	16	No	Flunarizine 5 mg/10 mg	Propranolol 80 mg /160 mg	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Diener, 2004 (24)	575	P	26	Yes	Topiramate 100 mg/200 mg	Propranolol 160	NA	Reduction in migraine attack frequency per month	TOP 100 = PROP 160 > placebo = TOP 200
Dodick, 2009 (25)	331	P	26	No	Topiramate 25–100 mg	Amitriptyline 25-100 mg	NA	Reduction in migraine attack frequency per month	Significant (?) change from baseline, No difference between treatments
Domingues, 2009 (26)	76	P	12	No	Propranolol 40 mg	Nortriptyline 20 mg	Propranolol 40 mg + nortriptyline 20 mg	Reduction in migraine attack frequency per month	Significant change from baseline for drug 1 and 3. No difference between treatments
Guidotti, 2007 (27)	38	P	2	No	Frovatriptan 2.5 mg	Transdermal oestrogen 25 microg	Naproxen sodium 500 mg	Perimenstrual incidence of migraine	Frovatriptan>naproxen = oestrogen
Gupta, 2007 (28)	60	C	16	Yes	Topiramate 25 mg	Lamotrigine 25 mg		Responder rate (50% decrease in mean migraine frequency/intensity).	Topiramate > lamotrigine = placebo
Israil, 2013 (44)	90	P	12	No	Amitriptyline 25 mg	Pizotifen 1.5 mg	Propranolol 20 mg	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Kalita, 2013 (45)	300	P	24	No	Amitriptyline 12.5–50 mg	Divalproate 250–1000 mg	NA	50% or greater reduction in monthly migraine attacks	Significant change from baseline, no difference between treatments
Keskinbora, 2008 (29)	73	P	12	No	Topiramate 25–200 mg	Amitriptyline 10–150 mg	Topiramate + amitriptyline combination	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Magalhães, 2010 (30)	72	P	13	No	Amitriptyline 25/50 mg	Botulinumtoxin 250 U	NA	50% or greater reduction in monthly migraine attacks	Significant change from baseline, no difference between treatments
Maizels, 2004 (31)	57	P	12	No	Magnesium 300 mg + riboflavin 400 mg + feverfew 100 mg	Riboflavin 25 mg	NA	50% or greater reduction in monthly migraine attacks	No difference between treatments
Mathew, 2009 (32)	60	P	11	No	OnabotulinumtoxinA (200 U avg)	Topiramate 100–200 mg	NA	The Treatment Responder Rate based on the Physician Global Assessment	Significant change from baseline, no difference between treatments
Millán-Guerrero, 2008 (33)	90	P	12	No	Subcutaneous histamine 1–10 ng twice a week	Topiramate 100 mg	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Millán-Guerrero, 2007 (34)	92	P	12	No	Subcutaneous histamine 1–10 ng twice a week	Sodium valproate 500 mg	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Millan-Guerrero, 2009 (35)	100	P	12	No	Subcutaneous histamine 1–10 ng twice a week	Botulinumtoxin A 50 U	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Mohammadianinejad, 2011 (41)	80	P	12	No	Zonisamide 200 mg	Topiramate 100 mg	NA	Reduction in migraine attack frequency per month	No difference between treatments
Schellenberg, 2008 (36)	30	P	16	No	Metoprolol 47,5-142.5 mg	Nebivolol 5 mg	NA	Reduction in migraine attack frequency per month	Significant? change from baseline, no difference between treatments
Shaygannejad, 2006 (37)	64	C	16	No	Topiramate 25–50 mg	Sodium valproate 200-400 mg	NA	Reduction in migraine attack frequency per month	Significant change from baseline, no difference between treatments
Stovner, 2014 (42)	72	C	12	Yes	Candesartan 16 mg	Propranolol 160 mg	NA	Reduction in migraine days per month	(Candesartan = propranolol) > placebo
Tarlaci, 2009 (38)	93	P	12	No	Venlafaxine (avg 72.8 mg)	Escitalopram (avg 12.4 mg)	NA	Not well-defined (headache intensity, frequency, duration, analgesic intake, and migraine disability assessment)	Significant change from baseline, no difference between treatments, except for the length of migraine attacks
Togha, 2008 (39)	125	P	12	No	Cinnarizine 75 mg	Sodium valproate 600 mg	NA	Reduction in migraine attack frequency per month	No difference between treatments
Wammes-van der Heijden, 2005 (40)	30	C	24	No	Acenocoumarol (INR 1.5–2.0)	Propranolol 80–160 mg	NA	Reduction in migraine attack frequency per month	No significant change from baseline, no difference between treatments
Zain, 2013 (46)	80	P	12	No	Topiramate 50–200 mg	Gabapentin 300–1200 mg	NA	Reduction in migraine attack frequency per month	No significant change from baseline, no difference between treatments

D: design; P: parallel group; C: cross-over; Dur: duration (weeks); Pla: placebo; INR: international normalised ratio.

Table 2.

Effect on migraine attacks of flunarizine 5 and 10 mg vs. propranolol for 16 weeks.

	Flunarizine 5 mg	Flunarizine 10 mg (5/7)^a	Propranolol 160 mg
Number of patients	263	275	270
Baseline mean frequency	3.1	3.0	3.0
Mean frequency during treatment	2.0	1.9	1.9
Mean change during treatment	1.1	1.1	1.1
% change	36%	37%	37%
Number completed	219 (83%)	222 (81%)	225 (83%)
Patients with a > 50% decrease in frequency	118/259 (46%)	141/264 (53%)	125/259 (48%)
Patients with AEs in the double-blind phase	88 (33%)	88 (32%)	88 (33%)
Withdrawal	Total: 142 (18%); due to AEs: 58 (7%)

Flunarizine 10 mg was administered only five out of seven days per week. AEs: adverse events.

Discussion

As is clear from this review of comparative, prophylactic RCTs in migraine from 2002 to 2014 most investigators and sponsors, and probably also ethical committees, prefer a trial design with a new drug and a standard drug but without placebo.

The ‘World Medical Association (WMA) Declaration of Helsinki: Ethical principles for medical research involving human subjects’ (47) is generally very restrictive on the use of placebo in RCTs (see Table 3). In contrast, the Drug Trial Subcommittee of the IHS recommends the use of placebo both when efficacy of a new prophylactic drug in migraine is to be demonstrated and when a new drug is compared with a ‘presumed standard drug’ (see Table 4) (13). WMA recommends that a new drug should, with very few exemptions (Table 3), be tested against ‘the best proven intervention(s)’. There is seemingly the assumption that ‘the best proven intervention’ is constantly effective in all performed RCTs with this treatment. The beta-blocker propranolol has for many years been considered ‘the standard proven drug’ for migraine prophylaxis. Numerous RCTs have, as mentioned in a recent Cochrane Review (48), been performed with propranolol: 26 RCTs compared propranolol with placebo, 13 RCTs compared propranolol with calcium antagonists, 10 RCTs compared propranolol with other beta-blockers, and 20 RCTs compared propranolol with other drugs. Also in the most recent trial program with topiramate propranolol was included as a comparator: topiramate vs. propranolol vs. placebo (24). The interesting point is the fact that propranolol was no better than placebo for migraine prophylaxis in seven of 26 (published) RCTs (48), whereas propranolol was significantly better than placebo in the meta-analysis of RCTs (48). Thus, when propranolol is used as the comparative standard drug with established efficacy in an RCT, one cannot predict whether the drug is actually effective in the RCT or whether an apparent effect of propranolol is due to a placebo effect, or the natural course of the disease, which can be up to, for example, 56% (12). There is thus a need to demonstrate in a comparative RCT in migraine prophylaxis that the ‘standard drug’ is effective if one is to evaluate correctly the performance of the ‘new drug’.

Table 3.

World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects (JAMA, 2013).

Use of placebo

The benefits, risks, burdens and effectiveness of a new intervention must be tested against those of the best proven intervention(s), except in the following circumstances:

• Where no proven intervention exists, the use of placebo, or no intervention, is acceptable; or

• Where for compelling and scientifically sound methodological reasons the use of any intervention less effective than the proven one, the use of placebo, or no intervention is necessary to determine the efficacy and safety of an intervention

• And the patients who receive any intervention less effective than the best proven one, placebo, or no intervention will not be subject to additional risks of serious or irreversible harm as result of not receiving the best proven intervention.

• Extreme care must be taken to avoid abuse of this option.

Table 4.

Citation from the International Headache Society’s guidelines for drug trials in migraine, Third Edition from 2012.

Design of prophylactic trials:

“Placebo control. Recommendations:

a. Drug used for migraine prophylaxis should be compared with placebo.

b. When two presumably active drugs are compared, placebo control should also be included in order to test the reactivity (assay sensitivity) of the trial, which would allow better generalizability of study results.

Comments: There is overwhelming evidence that migraineurs, similar to people with pain disorders, respond positively or negatively to placebo (placebo effect or response). Various studies have demonstrated that the placebo response rates (e.g. reduction in migraine frequency or 50% responder rate) in migraine prophylaxis usually range between 20% and 40%, or even higher. Therefore, the true drug effect cannot be established unless one can control for the placebo response. Also, the absence of a placebo arm in active-drug comparative trials does not allow a clear interpretation of active-drug effect, or lack thereof. For example if two active drugs are found equally effective in a non-placebo-controlled comparator trial, using conventional (as opposed to Bayesian) statistical approaches, the effect of either or both could simply be placebo-related. Finally, the use of non-contemporaneous placebo data is not valid scientifically and is highly discouraged in clinical research.

In addition, the choice of doses of the ‘standard drug’, in this case propranolol, varies considerably. Thus in four RCTs in the current review, propranolol doses were: 80 mg (17), 160 mg (23), 40 mg (26) and 80–160 mg (40). In the Cochrane Review propranolol doses of 60 mg to 320 mg were used in RCTs; in 49 of 55 RCTs doses of 120 mg or greater were used and in six RCTS only doses of 60–80 mg were used (48). Only one (49) of these six RCTs was placebo controlled. It was a cross-over study with response in 15/19 with propranolol vs. two of 19 with placebo. The Cochrane reviewer commented on this RCT: ‘small crossover trial with extremely positive results’ (48). This single, small RCT with propranolol 80 is not sufficient to establish small doses of propranolol as standard comparative drugs in prophylactic migraine RCTs. Thus if a non-established dose of an ‘established efficacious drug’ is used in comparative, prophylactic RCTs, it is not only unknown if that dose is effective in the trial population, but also if the dose has any migraine-prophylactic effect at all and therefore, again, placebo control is needed.

The largest prophylactic migraine RCT ever performed (23) is summarised in Table 1. The results exemplify the typical problems with performing even a very large comparative RCT without the inclusion of placebo. The stated aim of this trial ‘was to investigate whether the treatment of migraine could be improved by enhancement of the tolerability of flunarizine without compromising its efficacy as already established in comparison with propranolol’ (23). The basis for the stated presumed comparability between the two drugs was four RCTs (23,50 –53) in which propranolol 120–160 mg was found comparable to flunarizine 10 mg for migraine prophylaxis. It is noteworthy that in none of the four RCTs was placebo included. As shown in Table 1, the mean changes (1.1, 1.1, and 1.1) in migraine frequency during treatments (the primary efficacy parameter) were highly similar in the three treatment groups. Also the proportion of patients with adverse events (AEs) during the double-blind phase was remarkably similar (33%, 32%, and 33%). These results do not allow for any firm conclusion. Since it was not demonstrated in the statistical analyses that the three ‘active treatments’ were superior to placebo, the question of any efficacy demonstrated in this patient sample remains open. What was observed in the three treatment groups could merely be due to a placebo effect combined with the ‘natural course of the disease’ (12,54,55). Notably, in the Cochrane Review on propranolol (48) flunarizine and propranolol are described as being equipotent in migraine prophylaxis based on seven comparative RCTs without placebo control. Thus even these Cochrane reviewers seem unaware of the need for placebo control in comparative, prophylactic RCTs. Based on the 666 patients completing the study (see Table 1), we estimate that 75,000 patient days were, in our view, spent ‘in vain’ in this the largest ever prophylactic, migraine RCT (23) without placebo.

Generally, when comparing two drugs without placebo control, there are three possible outcomes: 1) ‘new drug’ is superior to ‘standard drug’, 2) ‘standard drug’ is superior to ‘new drug’, or 3) no significant difference between the drugs. Cases 1) and 2) are less problematic but, due to the highly variable treatment responses in migraine, there is a real risk that none of the drugs would have shown superiority to placebo in that particular trial. In case 3), efficacy of any of the drugs cannot be demonstrated without demonstrating the ‘reactivity of the sample” by including a placebo control. A similar decrease in frequency of migraine in the two treatment groups could, again, simply be due to ‘the natural course of the disease’ (54,55). Inability to demonstrate a statistically significant between-group difference could easily be due to high variability of the response within each group. Inclusion of a placebo arm could reveal superiority to placebo of one drug, but not the other. Therefore it cannot be concluded from a lack of significant difference between drugs that the drugs are ‘equally effective’. Also, when the aim is to show comparability, tolerability for the new drug compared to placebo is important for demonstrating whether the incidence of AEs with the new drug is higher or similar to placebo. In prophylactic RCTs of a single drug, placebo control is, for the same reasons, also crucial in order to determine if the drug has any effect apart from that of placebo. Finally, without the inclusion of placebo, in either case, the data cannot be used in meta-analyses and thus do not add to the scientific evidence on prophylactic treatment of migraine.

The WMA further recommends that any patients who do not receive the best proven treatment should not be subject to additional risks of serious or irreversible harms as a result of not receiving the best proven treatment (Table 3). The risk of such harms during placebo treatment in prophylactic RCTs in migraine is probably negligible since most patients, even up to six months, have a decrease in migraine frequency during placebo treatment (12), most likely partly due to ‘the natural course of the disease’, mentioned above (12,54,55).

In summary, the recommendations against placebo by the WMA, though likely relevant to studies of most disorders, do not apply to migraine prophylactic RCTs for two main reasons: a) migraine prophylactic treatment is characterised by an extremely variable placebo response, and b) placebo treatment in migraine does not cause serious or irreversible harm to patients.

On the contrary, in our view it is not ethical to include patients in RCTs that likely do not benefit actual or future patients, as is the case in 28 of 31 reviewed RCTs in the present study. The total number of days spent, in our opinion, ‘in vain’ by migraine patients in these 28 RCTs was approximately 360,000 and this high number could successfully be used when arguing with, for example, ethical committees for the use of placebo in comparative, prophylactic RCTs in migraine. The argument about the many patient days spent by migraine patients in badly designed RCTs is in our opinion a strong one, and it should be added to the more general arguments for placebo control in prophylactic migraine RCTs presented by the IHS (Table 4).

Clinical implications

In prophylactic migraine drug trials, the true drug effect cannot be established unless one can control for the placebo response.

Twenty-eight of 31 comparative prophylactic trials did not include placebo, rendering these studies largely inconclusive.

The total number of days spent by migraine patients in these 28 trials was approximately 360,000.

Investigators should include placebo in prophylactic migraine trials as recommended by the International Headache Society’s guidelines for drug trials in migraine.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Steiner

Stovner

Birbeck

. Migraine: The seventh disabler. J Headache Pain 2013; 14: 1–1.

Vos

Flaxman

Naghavi

. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990–2010: A systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012; 380: 2163–2196.

Evers

Afra

Frese

. EFNS guideline on the drug treatment of migraine – revised report of an EFNS task force. Eur J Neurol 2009; 16: 968–981.

Lipton

Bigal

Diamond

. Migraine prevalence, disease burden, and the need for preventive therapy. Neurology 2007; 68: 343–349.

Pringsheim

Davenport

Mackie

. Canadian Headache Society guideline for migraine prophylaxis. Canadian J Neurol Sci 2012; 39(2 Suppl 2): S1–S59.

Evers

Mylecharane

Nonsteroidal anti-inflammatory and miscellaneous drugs in migraine prophylaxis. In: JJ

Goadsby

Ramadan

(eds). The Headaches, Philadelphia: Lippincott Williams & Wilkins, 2006, pp. 553–566.

Silberstein

Tfelt-Hansen

Antiepileptic drugs in migraine prophylaxis. In: Olesen

Goadsby

Ramadan

(eds). The Headaches, Philadelphia: Lippincott Williams & Wilkins, 2006, pp. 545–551.

Tfelt-Hansen

Rolan

β-Adrenoceptor blocking drugs in migraine prophylaxis. In: Olesen

Goadsby

Ramadan

(eds). The Headaches, Philadelphia: Lippincott Williams & Wilkins, 2006, pp. 519–528.

Tfelt-Hansen

Saxena

Antiserotonin drugs in migraine prophylaxis. In: Olesen

Goadsby

Ramadan

(eds). The Headaches, Philadelphia: Lippincott Williams & Wilkins, 2006, pp. 529–537.

10.

Toda

Tfelt-Hansen

Calcium antagonists in migraine prophylaxis. In: Olesen

Goadsby

Ramadan

(eds). The Headaches, Philadelphia: Lippincott Williams & Wilkins, 2006, pp. 539–544.

11.

Tfelt-Hansen

Olesen

. Taking the negative view of current migraine treatments: The unmet needs. CNS Drugs 2012; 26: 375–382.

12.

Tfelt-Hansen

Hougaard

. Migraine: Differential effects of placebos in migraine clinical trials. Nat Rev Neurol 2014; 10: 10–11.

13.

Tfelt-Hansen

Pascual

Ramadan

. Guidelines for controlled trials of drugs in migraine: Third edition. A guide for investigators. Cephalalgia 2012; 32: 6–38.

14.

Hougaard

Tfelt-Hansen

. Are the current IHS guidelines for migraine drug trials being followed? J Headache Pain 2010; 11: 457–468.

15.

Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions Version 5.0.2, 2008.

16.

Ali

Awad

Al-Adl

. Efficacy of combined topiramate/thioctic acid therapy in migraine prophylaxis. Saudi Pharm J 2010; 18: 239–243.

17.

Ashtari

Shaygannejad

Akbari

. A double-blind, randomized trial of low-dose topiramate vs propranolol in migraine prophylaxis. Acta Neurol Scand 2008; 118: 301–305.

18.

Bartolini

Silvestrini

Taffi

. Efficacy of topiramate and valproate in chronic migraine. Clin Neuropharmacol 2005; 28: 277–279.

19.

Bavrasad

Nejad

Yarahmadi

. Assessment of the middle dose of topiramate in comparison with sodium valproate for migraine prophylaxis: A randomized double blind study. Int J Pharmacol 2010; 6: 670–675.

20.

Blumenfeld

Schim

Chippendale

. Botulinum toxin type A and divalproex sodium for prophylactic treatment of episodic or chronic migraine. Headache 2008; 48: 210–220.

21.

Bulut

Berilgen

Baran

. Venlafaxine versus amitriptyline in the prophylactic treatment of migraine: Randomized, double-blind, crossover study. Clin Neurol Neurosurg 2004; 107: 44–48.

22.

Cady

Schreiber

Porter

. A multi-center double-blind pilot comparison of onabotulinumtoxinA and topiramate for the prophylactic treatment of chronic migraine. Headache 2010; 51: 21–32.

23.

Diener

Matias-Guiu

Hartung

. Efficacy and tolerability in migraine prophylaxis of flunarizine in reduced doses: A comparison with propranolol 160 mg daily. Cephalalgia 2002; 22: 209–221.

24.

Diener

Tfelt-Hansen

Dahlöf

. Topiramate in migraine prophylaxis – results from a placebo-controlled trial with propranolol as an active control. J Neurol 2004; 251: 943–950.

25.

Dodick

Freitag

Banks

. Topiramate versus amitriptyline in migraine prevention: A 26-week, multicenter, randomized, double-blind, double-dummy, parallel-group noninferiority trial in adult migraineurs. Clin Ther 2009; 31: 542–559.

26.

Domingues

Silva

ALPD

Domingues

. A double-blind randomized controlled trial of low doses of propranolol, nortriptyline, and the combination of propranolol and nortriptyline for the preventive treatment of migraine. Arq Neuropsiquiatr 2009; 67: 973–977.

27.

Guidotti

Mauri

Barrilà

. Frovatriptan vs. transdermal oestrogens or naproxen sodium for the prophylaxis of menstrual migraine. J Headache Pain 2007; 8: 283–288.

28.

Gupta

Singh

Goyal

. Low-dose topiramate versus lamotrigine in migraine prophylaxis (the Lotolamp study). Headache 2007; 47: 402–412.

29.

Keskinbora

Aydinli

. A double-blind randomized controlled trial of topiramate and amitriptyline either alone or in combination for the prevention of migraine. Clin Neurol Neurosurg 2008; 110: 979–984.

30.

Magalhães

Menezes

Cardeal

. Botulinum toxin type A versus amitriptyline for the treatment of chronic daily migraine. Clin Neurol Neurosurg 2010; 112: 463–466.

31.

Maizels

Blumenfeld

Burchette

. A combination of riboflavin, magnesium, and feverfew for migraine prophylaxis: A randomized trial. Headache 2004; 44: 885–890.

32.

Mathew

Jaffri

SFA

. A double-blind comparison of onabotulinumtoxina (BOTOX) and topiramate (TOPAMAX) for the prophylactic treatment of chronic migraine: A pilot study. Headache 2009; 49: 1466–1478.

33.

Millán-Guerrero

Isais-Millán

Barreto-Vizcaíno

. Subcutaneous histamine versus topiramate in migraine prophylaxis: A double-blind study. Eur Neurol 2008; 59: 237–242.

34.

Millán-Guerrero

Isais-Millán

Barreto-Vizcaíno

. Subcutaneous histamine versus sodium valproate in migraine prophylaxis: A randomized, controlled, double-blind study. Eur J Neurol 2007; 14: 1079–1084.

35.

Millán-Guerrero

Isais-Millán

Barreto-Vizcaíno

. Subcutaneous histamine versus botulinum toxin type A in migraine prophylaxis: A randomized, double-blind study. Eur J Neurol 2009; 16: 88–94.

36.

Schellenberg

Lichtenthal

Wöhling

. Nebivolol and metoprolol for treating migraine: An advance on beta-blocker treatment? Headache 2008; 48: 118–125.

37.

Shaygannejad

Janghorbani

Ghorbani

. Comparison of the effect of topiramate and sodium valporate in migraine prevention: A randomized blinded crossover study. Headache 2006; 46: 642–648.

38.

Tarlaci

. Escitalopram and venlafaxine for the prophylaxis of migraine headache without mood disorders. Clin Neuropharmacol 2009; 32: 254–258.

39.

Togha

Mansoureh

Rahmat Jirde

. Cinnarizine in refractory migraine prophylaxis: Efficacy and tolerability. A comparison with sodium valproate. J Headache Pain 2008; 9: 77–82.

40.

Wammes-van der Heijden

Smidt

Tijssen

. Effect of low-intensity acenocoumarol on frequency and severity of migraine attacks. Headache 2005; 45: 137–143.

41.

Mohammadianinejad

Abbasi

Sajedi

. Zonisamide versus topiramate in migraine prophylaxis: A double-blind randomized clinical trial. Clin Neuropharmacol 2011; 34: 174–177.

42.

Stovner

Linde

Gravdahl

. A comparative study of candesartan versus propranolol for migraine prophylaxis: A randomised, triple-blind, placebo-controlled, double cross-over study. Cephalalgia 2014; 34: 523–532.

43.

Cosentino

Paladino

Maccora

. Efficacy and safety of topiramate in migraine prophylaxis: An open controlled randomized study comparing Sincronil and topamax formulations. Panminerva Med 2013; 55: 303–307.

44.

Israil

Ahmed

Rahman

. Efficacy of amitriptyline, pizotifen and propranolol in the prevention of migraine. Mymensingh Med J 2013; 22: 93–100.

45.

Kalita

Bhoi

Misra

. Amitriptyline vs divalproate in migraine prophylaxis: A randomized controlled trial. Acta Neurol Scand 2013; 128: 65–72.

46.

Zain

Khan

Alam

. Comparison of efficacy and safety of topiramate with gabapentin in migraine prophylaxis: Randomized open label control trial. J Pak Med Assoc 2013; 63: 3–7.

47.

World Medical Association. World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human subjects. JAMA 2013; 310: 2191–2194.

48.

Linde

Rossnagel

. Propranolol for migraine prophylaxis. Cochrane Database Syst Rev 2004; 2: CD003225–CD003225.

49.

Weber

Reinmuth

. The treatment of migraine with propranolol. Neurology 1972; 22: 366–369.

50.

Gawel

Kreeft

Nelson

. Comparison of the efficacy and safety of flunarizine to propranolol in the prophylaxis of migraine. Can J Neurol Sci 1992; 19: 340–345.

51.

Ludin

. Flunarizine and propranolol in the treatment of migraine. Headache 1989; 29: 219–224.

52.

Lücking

Oestreich

Schmidt

. Flunarizine vs propranolol in the prophylaxis of migraine: Two double-blind comparative studies in more than 400 patients. Cephalalgia 1988; 8(Suppl 8): 21–26.

53.

Shimell

Fritz

Levien

. A comparative trial of flunarizine and propranolol in the prevention of migraine. S Afr Med J 1990; 77: 75–77.

54.

Olesen

Krabbe

Tfelt-Hansen

. Methodological aspects of prophylactic drug trials in migraine. Cephalalgia 1981; 1: 127–141.

55.

Krogsbøll

Hróbjartsson

Gøtzsche

. Spontaneous improvement in randomised clinical trials: Meta-analysis of three-armed trials comparing no treatment, placebo and active intervention. BMC Med Res Methodol 2009; 9: 1–1.