Clinical Trial

published: 11 October 2017

doi: 10.3389/fneur.2017.00524

The Efficacy of Venlafaxine,

Flunarizine, and Valproic acid in the
Prophylaxis of Vestibular Migraine
Fenye Liu 1, Tianbao Ma 2, Xiaolin Che 3, Qirong Wang 3 and Shudong Yu 3*

 Department of Traditional Chinese Medicine, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, China,
1

 Department of Hearing Central, Women and Children’s Health Care Hospital of Linyi, Linyi, China, 3 Department of
2

Otolaryngology, Shandong Qianfoshan Hospital, Jinan, China

Background: Different types of medications are currently used in vestibular migraine

(VM) prophylaxis, although recommendations for use are generally based on expert
opinion rather than on solid data from randomized trials. We evaluated the efficacy and
safety of venlafaxine, flunarizine, and valproic acid in a randomized comparison trial for
VM prophylaxis.
Methods: Subjects were randomly allocated to one of three groups (venlafaxine group,
flunarizine group, and valproic acid group). To assess the efficacy of treatment on vertigo
symptoms, the following parameters were assessed at baseline and 3  months after
Edited by: treatment: Dizziness Handicap Inventory (DHI) scores, number of vertiginous attacks
Jeffrey P. Staab,
Mayo Clinic, United States in the previous month, and Vertigo Severity Score (VSS). Adverse events also were
Reviewed by:
evaluated.
Alexandre Bisdorff,
Centre Hospitalier Emile Mayrisch,
results: A decrease in DHI total scores was shown following treatment with all three
Luxembourg medications, with no obvious differences between the groups. Treatment effects differed,
Nicolas Perez,
however, in the DHI physical, functional, and emotional domains with only venlafaxine
Universidad de Navarra, Spain
showing a decreased effect in all of three domains. Flunarizine and valproic acid showed
*Correspondence:
Shudong Yu an effect in only one DHI domain. Venlafaxine and flunarizine showed decreased VSS
[email protected] scores (p = 0 and p = 0.03, respectively). Although valproic acid had no obvious effect
on VSS (p  =  0.27), decreased vertigo attack frequency was observed in this group
Specialty section:
This article was submitted to
(p = 0). Venlafaxine also had an effect on vertigo attack frequency (p = 0), but flunarizine
Neuro-Otology, had no obvious effect (p = 0.06). No serious adverse events were reported in the three
a section of the journal
groups.
Frontiers in Neurology

Received: 02 June 2017 Conclusion: Our data confirm the efficacy and safety of venlafaxine, flunarizine,
Accepted: 20 September 2017 and valproic acid in the prophylaxis of VM, venlafaxine had an advantage in terms of
Published: 11 October 2017
emotional domains. Venlafaxine and valproic acid also were shown to be preferable
Citation:
Liu F, Ma T, Che X, Wang Q and Yu S to flunarizine in decreasing the number of vertiginous attacks, but valproic acid was
(2017) The Efficacy of Venlafaxine, shown to be less effective than venlafaxine and flunarizine to decrease vertigo severity.
Flunarizine, and Valproic Acid in the
Prophylaxis of Vestibular Migraine. Trial registration: ChiCTR-OPC-17011266 (http://www.chictr.org.cn/).
Front. Neurol. 8:524.
doi: 10.3389/fneur.2017.00524 Keywords: vestibular migraine, prophylaxis, efficacy, venlafaxine, flunarizine, valproic acid

Frontiers in Neurology  |  www.frontiersin.org 1 October 2017 | Volume 8 | Article 524

Liu et al. Three Medicines in the Prophylaxis of VM

INTRODUCTION TABLE 1 | Diagnostic criteria for definite and probable VM (pVM).

1. Definite VM
Migraine and vertigo are among the most common health con- A. At least 5 episodes with vestibular symptoms of moderate or severe
cerns in the general population. The link between migraine and intensity, lasting 5 min to 72 h
vertigo was described for the first time by Aretaeus of Cappadocia B. Current or previous history of migraine with or without aura according to
in 131 BC. Nowadays, vestibular migraine (VM) is considered the International Classification of Headache Disorders (ICHD)
C. One or more migraine features with at least 50% of the vestibular
a distinct diagnostic entity by both the Barany Society and the episodes:
International Headache Society (1–3), and VM is considered one – headache with at least two of the following characteristics: one
of the most common vestibular disorders in the general popula- sided location, pulsating quality, moderate or severe pain intensity,
tion with a lifetime prevalence of 1% and 1-year prevalence of aggravation by routine physical activity
– photophobia and phonophobia
0.9% (4).
– visual aura
Vestibular migraine presents with attacks of vertigo or dizzi- D. Not better accounted for by another vestibular or ICHD diagnosis
ness lasting several minutes to 3  days. Vertigo can precede the 2. pVM
migraine attack but can also occur during or after the headache A. At least 5 episodes with vestibular symptoms of moderate or severe
(5, 6). Vertigo can also occur in the absence of a typical migrainous intensity, lasting 5 min to 72 h
B. Only one of the criteria B and C for vestibular migraine is fulfilled
headache, and symptoms such as phono-/photophobia, osmo-
(migraine history or migraine features during the episode)
phobia, nausea, and vomiting, and aggravation by movement C. Not better accounted for by another vestibular or ICHD diagnosis
can present in some patients. Currently, there are no biological
Vestibular symptoms include the following: spontaneous vertigo, positional vertigo,
markers for the diagnosis of VM. Moreover, VM vertigo shares visually induced vertigo, head motion-induced vertigo, and head motion-induced
features with some other clinical conditions (7). Taken together, dizziness with nausea. Vestibular symptoms are “moderate” if they interfere with but
these factors can present diagnostic difficulties. do not prohibit daily activities and “severe” if patients cannot continue daily activities.
Since the mechanisms underlying VM remain insufficiently Adapted from Lempert et al. (1–3).

known, different types of medications are typically used in VM

prophylaxis including β-blockers (propranolol or metoprolol), they took throughout the trial. To assess the efficacy of treatment
calcium overload blockers (flunarizine), anticonvulsants (valp- on vertigo symptoms, the following parameters were assessed
roic acid or topiramate), and antidepressants (amitriptyline and at baseline and 3  months after treatment: Dizziness Handicap
venlafaxine) (8–11). Unfortunately, current treatment recom- Inventory (DHI) scores (13), number of vertiginous attacks in
mendations are based on expert opinion rather than on solid the previous month, and Vertigo Severity Score (VSS). These
data from randomized trials. Few studies to date have evaluated primary outcome measure methods were similar to those recom-
the effectiveness of venlafaxine therapy in VM prophylaxis (12). mended for migraine (14). The DHI is a validated, self-reported
In this randomized comparison trial study, the efficacy and safety questionnaire designed to evaluate the precipitating physical
of venlafaxine, flunarizine, and valproic acid in VM prophylaxis factors associated with dizziness and unsteadiness as well as the
were investigated. functional and emotional consequences of vestibular disease. VSS
is a visual analog scale with 0 indicating the absence of vertigo and
METHODS 10 representing extremely severe vertigo. Adverse events related
to study medication were assessed during each visit (Figure 1).
Patients
A single-blinded randomized comparison trial of 3 months dura- Treatment
tion was carried out in Shandong Qianfoshan Hospital according Patients in the VG received 25 mg venlafaxine (Efexor XR, 75 mg,
to a protocol approved by the hospital ethics committee. Patients Wyeth, USA) at bedtime for 6 days, followed to a total daily dose
were enrolled between January 1, 2016 and December 25, 2016. An of 37.5 mg (low than normal dosage). Patients in the flunarizine
ear, nose, throat, and neurotological examination was performed group received 10 mg flunarizine (Sibelium, 5 mg, Xian Janssen,
followed by specific audiovestibular investigations and imaging China) at bedtime for the entire treatment period. Patients in the
when required. Eligible participants included male and female VAG received valproic acid (Depakin, 500  mg, Sanofi, France)
outpatients aged 18  years or older with a primary diagnosis of at a dose of 1,000 mg/day (twice daily) for the entire treatment
VM [confirmed VM or probable VM (pVM)] based on standard period.
criteria (Table 1) (1–3). Additional inclusion criteria were two or
more migraine attacks per month or very disabling attacks and no Statistical Analysis
improvement observed following non-pharmaceutical treatment. Data were analyzed using the SPSS version 19 statistical pack-
Patients with associated benign paroxysmal positional vertigo, age (IBM Corporation, New York City, NY, USA). Baseline
Meniere’s disease, chronic discharging ear, history of ear surgery, demographics and clinical characteristics were compared using
profound hearing loss, women who were breastfeeding, and the ANOVA test for numeric data (age) and chi-square tests for
patients with major cardiovascular, metabolic, gastrointestinal, nominal data (sex and diagnosis). Paired t-tests were used to
and neurologic diseases were excluded from the study. Subjects compare VSS, DHI, and attack frequency before and after treat-
were randomly allocated to one of three groups [venlafaxine group ment. ANOVA tests were used to compare changes in DHI scores
(VG), flunarizine group, and valproic acid group (VAG)] by the from baseline to the final visit. Significance was set at p < 0.05 for
order of visits. The patients were unaware of the medication that all statistical tests.

Frontiers in Neurology  |  www.frontiersin.org 2 October 2017 | Volume 8 | Article 524

Liu et al. Three Medicines in the Prophylaxis of VM

TABLE 2 | Demographic profile of patients in the study.

Venlafaxine Flunarizine Valproic acid p-Value

Sex
Male 7 8 5
Female 16 14 15 0.727
Total 23 22 20
Diagnosis
dVM 6 8 6 0.754
pVM 17 14 14
Age (years)
( x ± S) 53.22 ± 15.55 51.45 ± 15.43 52.35 ± 16.01 0.931

dVM, definite VM; pVM, probable VM.

decrease was observed after treatment with all three medica-

tions (p < 0.05) although there were no differences between the
groups (p  >  0.05, Figure  2). Vertigo severity scoring on VSS
was inconsistent after treatment in the three groups, although
venlafaxine and flunarizine showed decreased VSS scores (p = 0
and p = 0.03, respectively). Although valproic acid had no obvi-
ous effect on VSS (p = 0.27), decreased vertigo attack frequency
was observed in this group (p = 0). Venlafaxine had an effect on
vertigo attack frequency (p = 0), but flunarizine had no obvious
effect (p = 0.06).
FIGURE 1 | CONSORT flow diagram of patient disposition.
Safety
No serious side effects were observed in patients who completed
RESULTS the study period. In the VG, two patients suffered from nausea,
one suffered from insomnia, one suffered from palpitation, and
Patients one complained of lethargy. In the VAG, two patients reported
A total of 75 subjects with either confirmed VM or pVM were nausea, one reported somnolence, and one reported indigestion.
recruited into this study and were treated with venlafaxine, In the flunarizine group, five patients reported somnolence and
flunarizine, or valproic acid for 3 months. Five patients in the one patient reported nausea.
VAG (unexpected improvement, three patients; adverse events,
two patients), two patients in the venlafaxine group (consent
withdrawn), and three patients in the flunarizine group (lost
DISCUSSION
to follow-up, two patients; consent withdrawn, one patient) Prophylactic treatment of VM is mainly intended to reduce
terminated the study prematurely (Figure 1). Data on the age, attack frequency, duration, and severity of vertigo. Among the
gender, and diagnosis in the different groups are shown in many types of medications prescribed for migraine prophylaxis,
Table 2. There were no conspicuous differences between three valproic acid, venlafaxine, and flunarizine are the most repre-
groups at baseline. Data on the VSS, DHI, and attack frequency sentative. In this randomized comparison trial, the efficacy and
in the different groups before treatment are shown in Table 3; safety of these three medications were investigated for prophy-
no conspicuous differences were observed between the three laxis in patients diagnosed with VM. Our findings suggest that
groups (p > 0.05). the medications were safe and effective in reducing DHI total
score but had difference effects in the domains of VSS score
Efficacy and attack frequency; venlafaxine had an advantage in terms
Changes in the different outcome measures in both groups are of emotional domains. Venlafaxine and valproic acid also were
shown in Table  3 and Figure  2. Significantly decreased total shown to be preferable to flunarizine in decreasing the number
DHI scores were observed following treatment with all three of vertiginous attacks, but valproic acid was shown to be less
medications (p  <  0.05, Table  3) with no obvious differences effective than venlafaxine and flunarizine to decrease vertigo
between three groups (p > 0.05, Figure 2). However, treatment severity.
effects differed according to the DHI physical, functional, and Although several studies have reported the efficacy of venla-
emotional domains. The emotional domains of DHI were signif- faxine in the prophylaxis of migraine (15–17), very little data exist
icantly decreased following treatment with venlafaxine (p = 0), on its efficacy for VM prophylaxis (12) although some studies
while no difference was observed in the flunarizine and VAGs have investigated flunarizine (16, 18) and valproic acid for VM
(p  =  0.12 and p  =  0.11, respectively). A significant difference prophylaxis (9, 19). In this study, all three medications had effects
was observed between the three groups (p  <  0.05, Figure  2). on VM. Valproic acid had no effect on the severity of vertigo
For the physical and functional domains of DHI, a significant although it significantly reduced the number of vertigo episodes.

Frontiers in Neurology  |  www.frontiersin.org 3 October 2017 | Volume 8 | Article 524

Liu et al. Three Medicines in the Prophylaxis of VM

TABLE 3 | Groups before and after treatment with regard to symptom (VSS), disability (DHI), and attack frequency.

Venlafaxine Flunarizine Valproic acid

Before After p Before After p Before After p

VSS 5.96 (1.72) 3.78 (1.28) 0 6.41 (1.99) 5.86 (1.55) 0.03 5.8 (1.82) 5.3 (1.08) 0.27
DHI-e 14.17 (6.95) 9.39 (5.03) 0 16.91 (6.84) 15.45 (7.07) 0.115 16.8 (4.7) 15.3 (4.91) 0.11
DHI-f 14.87 (7.11) 11.57 (5.36) 0.01 16.64 (5.57) 13.45 (6.24) 0.013 16.60 (5.20) 13.2 (4.96) 0.02
DHI-p 12.61 (4.80) 9.91 (5.24) 0.018 13.00 (4.44) 10.91 (4.69) 0.041 13.4 (4.45) 10.30 (4.91) 0.01
DHI-t 41.74 (16.90) 31.3 (14.14) 0.001 46.64 (15.15) 39.82 (16.35) 0.019 46.80 (13.45) 38.7 (13.58) 0.02
Frequency 5.83 (3.2) 3.09 (1.68) 0 4.95 (3.28) 4.15 (2.46) 0.057 5.1 (3.14) 2.35 (1.79) 0

Values are expressed as the median (SD).

VSS, Vertigo Severity Score; DHI-t, Dizziness Handicap Inventory-total score; e, emotional; f, functional; p, physical; DHI, Dizziness Handicap Inventory.

venlafaxine confers some advantages over other medications for

VM prophylaxis through several factors. Similar to the onset of
migraine, 5-HT levels affect the onset of vestibular symptoms
(24, 25). Venlafaxine decreases 5-HT levels and subsequently
vertigo attack frequency and can also reduce the levels of certain
inflammatory cytokines (26–28) that play a role in episodic
vertigo. Venlafaxine also exerts neuroprotection effects (29) that
may decrease the severity of vertigo.
It should be noted that the dose of venlafaxine used in our
study was lower than that reported previously (12). In our pre-
liminary study, where we found that a daily dose of 37.5 mg had
similar effects to the maximum daily dose of 75 mg. Therefore, we
used 37.5 mg as the maximum daily dose in this study to further
reduce side effects. In this study, none of the three medications
was associated with serious side effects, and so was considered
safe at the investigated doses. It also should be noted that most of
FIGURE 2 | Change in DHI scores from baseline to final visit for three the patients in our study were pVM. The reason is that pVM has
medications ( x ± SE ). DHI, Dizziness Handicap Inventory; VG, venlafaxine higher morbidity than definite VM.
group; FG, flunarizine groups; VAG, valproic acid group (#p > 0.05; The results of this study should be interpreted within its
*p < 0.05). limitations. This study had no placebo arm because the major
purpose was to investigate the efficacy of three recommendatory
medications (10, 12, 18, 30). But that does not allow conclusions
Valproic acid is an anticonvulsant medication, and its mechanism to be drawn on the absolute efficacy of the medications. Second,
of migraine prophylaxis may be related to ion channel blocking because both drugs showed beneficial effects and the effects
(20). Flunarizine is also a channel blocker, selectively blocking depended on subjective experience or feeling (the outcome
calcium channels. Therefore, ion channels may have a role in the measures used mostly based on patients’ feeling), it was unclear
occurrence of VM; calcium channels may affect the severity of whether the results were placebo effects. The low patient numbers
vertigo while other ion channels that can be blocked by valproic may also have provided less power to detect between-group
acid may affect attack frequency. differences. Furthermore, the efficacy of the three medications
Venlafaxine, a serotonin–norepinephrine reuptake inhibitor, was examined for only a short period of time. Future long-term
is a clinically effective and safe medication that is widely used studies with a larger sample size and placebo arm are therefore
to treat depression. In recent years, it has been shown to have warranted.
potential in migraine prophylaxis (15–17). Similar to its use in
the treatment of migraine, venlafaxine is a safe and effective
medication for the prophylaxis of VM. It is well known that CONCLUSION
the link between vertigo and anxiety, VM patients were more
anxious than migraine patients without vestibular symptoms Our data confirm the efficacy and safety of venlafaxine, flunar-
(21, 22). As a medication to treat anxiety, it has obvious izine, and valproic acid in the prophylaxis of VM. Venlafaxine
effects on emotional symptom relief that differ from the other had an advantage in terms of emotional domains. Venlafaxine
medications investigated in the current study. For this reason, and valproic acid also were shown to be preferable to flunarizine
venlafaxine has a better response in VM patients, especially in decreasing the number of vertiginous attacks, but valproic acid
those with anxiety (23). In our study, venlafaxine can also was shown to be less effective than venlafaxine and flunarizine to
decrease vertigo attack frequency and severity. We believe that decrease vertigo severity.

Frontiers in Neurology  |  www.frontiersin.org 4 October 2017 | Volume 8 | Article 524

Liu et al. Three Medicines in the Prophylaxis of VM

ETHICS STATEMENT ACKNOWLEDGMENTS

The study was approved by the ethics committee of Shandong The authors would like to thank all the patients.
Qianfoshan Hospital. All patients provided written informed
consent. FUNDING
AUTHOR CONTRIBUTIONS This study was supported by the Shandong Science &
Technology development program of TCM (grant number
SY conceived and designed the study. FL, TM, and XC performed 2013-195) and Shandong Science & Technology develop-
the experiments. FL wrote the paper. QW reviewed and edited the ment program of medicine and health (grant number
manuscript. All the authors read and approved the manuscript. 2015WS0246).

REFERENCES 17. Ozyalcin SN, Talu GK, Kiziltan E, Yucel B, Ertas M, Disci R. The efficacy
and safety of venlafaxine in the prophylaxis of migraine. Headache (2005)
1. Lempert T, Olesen J, Furman J, Waterston J, Seemungal B, Carey J, et  al. 45:144–52. doi:10.1111/j.1526-4610.2005.05029.x
Vestibular migraine: diagnostic criteria: consensus document of the Bárány 18. Taghdiri F, Togha M, Razeghi JS, Refaeian F. Cinnarizine for the prophylaxis of
Society and the International Headache Society. Nervenarzt (2013) 84:511–6. migraine associated vertigo: a retrospective study. Springerplus (2014) 3:231.
doi:10.1007/s00115-013-3768-x doi:10.1186/2193-1801-3-231
2. Lempert T, Olesen J, Furman J, Waterston J, Seemungal B, Carey J, et  al. 19. Celiker A, Bir LS, Ardiç N. Effects of valproate on vestibular symptoms and
Vestibular migraine: diagnostic criteria. Consensus document of the Barany electronystagmographic findings in migraine patients. Clin Neuropharmacol
Society and the International Headache Society. Rev Neurol (Paris) (2014) (2007) 30:213–7. doi:10.1097/wnf.0b013e31803bb3ee
170:401–6. doi:10.1016/j.neurol.2013.05.010 20. Hoffmann J, Akerman S, Goadsby PJ. Efficacy and mechanism of anticonvul-
3. Lempert T, Olesen J, Furman J, Waterston J, Seemungal B, Carey J, et  al. sant drugs in migraine. Expert Rev Clin Pharmacol (2014) 7:191–201. doi:10.
Vestibular migraine: diagnostic criteria. J Vestib Res (2012) 22:167–72. 1586/17512433.2014.885835
doi:10.3233/VES-2012-0453 21. Lahmann C, Henningsen P, Brandt T, Strupp M, Jahn K, Dieterich M, et al.
4. Neuhauser HK, Radtke A, von BM, Feldmann M, Lezius F, Ziese T, et  al. Psychiatric comorbidity and psychosocial impairment among patients
Migrainous vertigo: prevalence and impact on quality of life. Neurology (2006) with vertigo and dizziness. J Neurol Neurosurg Psychiatry (2015) 86:302–8.
67:1028–33. doi:10.1212/01.wnl.0000237539.09942.06 doi:10.1136/jnnp-2014-307601
5. Cutrer FM, Baloh RW. Migraine-associated dizziness. Headache (1992) 22. Kutay Ö, Akdal G, Keskinoğlu P, Balcı BD, Alkın T. Vestibular migraine
32:300–4. doi:10.1111/j.1526-4610.1992.hed3206300.x patients are more anxious than migraine patients without vestibular symp-
6. Neuhauser H, Leopold M, von BM, Arnold G, Lempert T. The interrelations toms. J Neurol (2017) 5:1–5. doi:10.1007/s00415-017-8439-6
of migraine, vertigo, and migrainous vertigo. Neurology (2001) 56:436–41. 23. Staab JP. Clinical clues to a dizzying headache. J Vestib Res (2011) 21:331–40.
doi:10.1212/WNL.56.4.436 doi:10.3233/VES-2011-0426
7. Fasunla AJ, Ibekwe TS, Nwaorgu OG. Migraine-associated vertigo: a review 24. Ahn SK, Balaban CD. Distribution of 5-HT1B and 5-HT1D receptors in the
of the pathophysiology and differential diagnosis. Int J Neurosci (2012) inner ear. Brain Res (2010) 1346:92–101. doi:10.1016/j.brainres.2010.05.057
122:107–13. doi:10.3109/00207454.2011.631717 25. Khalmuratova R, Hah YS, Ahn SK, Jeon SY, Kim JP, Balaban CD.
8. Verspeelt J, De Locht P, Amery WK. Postmarketing study of the use of flu- Immunohistochemical and biomolecular identification of 5-HT7 receptor in
narizine in vestibular vertigo and in migraine. Eur J Clin Pharmacol (1996) rat vestibular nuclei. J Vestib Res (2010) 20:401–6. doi:10.3233/VES-2010-0390
51:15–22. doi:10.1007/s002280050154 26. Espinosa-Sanchez JM, Lopez-Escamez JA. New insights into pathophysiology
9. Bisdorff AR. Management of vestibular migraine. Ther Adv Neurol Disord of vestibular migraine. Front Neurol (2015) 6:12. doi:10.3389/fneur.2015.00012
(2011) 4:183–91. doi:10.1177/1756285611401647 27. Hajhashemi V, Minaiyan M, Banafshe HR, Mesdaghinia A, Abed A.
10. Obermann M, Strupp M. Current treatment options in vestibular migraine. The anti-inflammatory effects of venlafaxine in the rat model of carrageenan-
Front Neurol (2014) 5:257. doi:10.3389/fneur.2014.00257 induced paw edema. Iran J Basic Med Sci (2015) 18:654–8.
11. Maldonado FM, Birdi JS, Irving GJ, Murdin L, Kivekas I, Strupp M. 28. Lee YY, Yang YP, Huang PI, Li WC, Huang MC, Kao CL, et al. Exercise sup-
Pharmacological agents for the prevention of vestibular migraine. Cochrane presses COX-2 pro-inflammatory pathway in vestibular migraine. Brain Res
Database Syst Rev (2015) 6:CD010600. doi:10.1002/14651858.CD010600. Bull (2015) 116:98–105. doi:10.1016/j.brainresbull.2015.06.005
pub2 29. Zepeda R, Contreras V, Pissani C, Stack K, Vargas M, Owen GI, et al. Venlafaxine
12. Salviz M, Yuce T, Acar H, Karatas A, Acikalin RM. Propranolol and venla- treatment after endothelin-1-induced cortical stroke modulates growth factor
faxine for vestibular migraine prophylaxis: a randomized controlled trial. expression and reduces tissue damage in rats. Neuropharmacology (2016)
Laryngoscope (2016) 126:169–74. doi:10.1002/lary.25445 107:131–45. doi:10.1016/j.neuropharm.2016.03.011
13. Jacobson GP, Newman CW. The development of the dizziness handicap 30. Stolte B, Holle D, Naegel S, Diener HC, Obermann M. Vestibular migraine.
inventory. Arch Otolaryngol Head Neck Surg (1990) 116:424–7. doi:10.1001/ Cephalalgia (2015) 35:262–70. doi:10.1177/0333102414535113
archotol.1990.01870040046011
14. Tfelt-Hansen P, Block G, Dahlöf C, Diener HC, Ferrari MD, Goadsby PJ, Conflict of Interest Statement: The authors declare that they have no financial and
et  al. Guidelines for controlled trials of drugs in migraine: second personal relationships with other people or organizations that can inappropriately
edition. Cephalalgia (2000) 20:765–86. doi:10.1046/j.1468-2982.2000. influence our work, and there is no professional or other personal interest of any
00117.x nature or kind in any product, service, and/or company that could be construed
15. Adelman LC, Adelman JU, Von Seggern R, Mannix LK. Venlafaxine extended as influencing the position presented in, or the review of, the manuscript entitled.
release (XR) for the prophylaxis of migraine and tension-type headache:
a retrospective study in a clinical setting. Headache (2000) 40:572–80. Copyright © 2017 Liu, Ma, Che, Wang and Yu. This is an open-access article dis-
doi:10.1046/j.1526-4610.2000.00089.x tributed under the terms of the Creative Commons Attribution License (CC BY).
16. Bulut S, Berilgen MS, Baran A, Tekatas A, Atmaca M, Mungen B. Venlafaxine The use, distribution or reproduction in other forums is permitted, provided the
versus amitriptyline in the prophylactic treatment of migraine: randomized, original author(s) or licensor are credited and that the original publication in this
double-blind, crossover study. Clin Neurol Neurosurg (2004) 107:44–8. journal is cited, in accordance with accepted academic practice. No use, distribution
doi:10.1016/j.clineuro.2004.03.004 or reproduction is permitted which does not comply with these terms.

Frontiers in Neurology  |  www.frontiersin.org 5 October 2017 | Volume 8 | Article 524