Evaluation of the efficacy of memantine in the treatment of fibromyalgia: study protocol for a doubled-blind randomized controlled trial with six-month follow-up.

0.29 MB·English

by  Olivan-Blazquez, Barbara

Olivan-Blázquezetal.Trials2013,14:3
http://www.trialsjournal.com/content/14/1/3 TRIALS
STUDY PROTOCOL Open Access
Evaluation of the efficacy of memantine in the
treatment of fibromyalgia: study protocol for a
doubled-blind randomized controlled trial with
six-month follow-up
Bárbara Olivan-Blázquez1,2, Marta Puebla2, Bárbara Masluk2, Mari-Cruz Pérez-Yus2, Raquel Arcega2, Eva Andrés3,
Yolanda López-del-Hoyo1,2, Rosa Magallon2,4, Miquel Roca2,5 and Javier Garcia-Campayo2,6,7*
Abstract
Background: Fibromyalgia is a prevalent chronic rheumatic disease ofgreat clinical importance. Recent studies
have found raised levels of glutamate in theinsula, hippocampus and posterior cingulate cortex regions ofthe
brains offibromyalgia (FM) patients. This finding has led researchers to speculate about theusefulness of
glutamate-blocking drugs such as memantineinthetreatment of fibromyalgia. The hypothesis ofthis study is that
theadministration ofmemantinewill reduce theglutamate levels, and futhermore, will decrease the perceived
pain. The aim ofthis study is to evaluate theefficacy of memantineinthe treatment of pain (pain perception). A
secondary objective is to evaluate the efficacy of memantine in thetreatmentof other clinical symptoms of FM,
and to evaluate theefficacy of memantine inreducing brainlevels of glutamate, and itseffects on thecentral
nervous system as a whole.
Method/Design: Adouble-blind parallel randomizedcontrolled trial. Participants, Seventy patients diagnosed with
FM will be recruited from primary health care centers inZaragoza, Spain.Intervention. The subjects will be
randomized intwo groups: A) A treatment group (n= 35),which will receive 20 mg ofmemantinedaily; B) A
control group (n = 35),to which will be administered a placebo. There will be a six-monthfollow-upperiod
(including a titration period of one month). Outcomes. The main efficacy variable of this study is pain (pain
perception). The secondaryefficacy variablesare clinical symptoms (pain threshold, cognitivefunction,health status,
anxiety, depression, clinical impressionand qualityof life) and glutamate levels indifferent regions ofthe brain,
which will be assessed bymagneticresonance spectroscopy.Randomization and blinding. Randomization has been
computer-generated, and the random allocation sequencewill be implemented by telephone. Subjectsof the
study and theresearch assistants will be blinded to group assignment.
Discussion: There is a need for thedevelopment ofinnovativeand more effective treatments for fibromyalgia. This
clinical trial will determine whether memantinecan be an effective pharmacological treatment for fibromyalgia
patients.
Trial registration: CurrentControlled Trials ISRCTN45127327 EUDRACT 2011-006244-73
Keywords: Fibromyalgia, Memantine, Chronic pain, Magnetic resonance spectroscopy, Randomized controlled trial
*Correspondence:[email protected]
2ReddeActividadesPreventivasydePromocióndelaSalud(REDIAPP)
(RD06/0018),InstitutoAragonésdeCienciasdelaSalud(IACS),Aragón,Spain
6ServiciodePsiquiatría,HospitalMiguelServetyUniversidaddeZaragoza,
Zaragoza,Spain
Fulllistofauthorinformationisavailableattheendofthearticle
©2013Olivan-Blázquezetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsofthe
CreativeCommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,
distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
Olivan-Blázquezetal.Trials2013,14:3 Page2of9
http://www.trialsjournal.com/content/14/1/3
Background presents significant difficulties [17]. The NMDA receptor
Fibromyalgia (FM) is achronic rheumatic disease of cur- antagonistmemantineisaderivativeofamantadine,adrug
rently unknown aetiology that is characterized by the whichhasbeenusedtotreatParkinson’sdisease,spasticity,
presenceofdiffusemusculoskeletalpainandpainfulsen- convulsions, vascular dementia and Alzheimer's disease,
sitivitytotouchinatleast11of18definedtriggerpoints and has an excellent clinical safety record spanning more
[1]. The prevalence of this syndrome in Europe is cal- than20years.Itisanon-competitiveopen-channelblocker
culated to be approximately 2.9% (95% CI: 2.4 to 3.4) that dissociates from the channel, which allows it to limit
[2].TheprevalenceofFMinrheumatologyconsultations the pathological activity of the NMDA receptor without
in Spain was found to be 12% (2.2% in men and 15.5% affectingnormalsynapticactivity[18].
in women) [3]. Because of its high prevalence, there is Memantine has shown a very low incidence of side
great clinical impact on patients in terms of disability effects in clinical trials on humans [19,20], while a re-
and loss of quality of life, and significant health care cent extension of trials has demonstrated the drug’s
costs. FM treatments are believed to have limited effi- clinical tolerability, even with prolonged use [21]. Simi-
cacy, withan effect sizeofapproximately 0.5[4]. lar low rates of adverse effects are expected with the
Pain is the most common and debilitating symptom of treatment with memantine of other disorders such as
FM. The cause of hyperalgesia in FM is poorly under- pain [22,23], migraine [24] and fibromyalgia [25]. The
stood. It is suspected that there is an alteration of the clinically approved dose of memantine for humans
function of structures of the central nervous system. In starts with 5 mg/day, increasing progressively over a
recent years, the neurophysiology of the phenomenon of period of several weeks to 20 mg/day. This progressive
pain has led to increased interest in employing different dose adjustment may contribute to the drug’s lack of
neuroimaging methods such as positron emission to- side effects [19]. While this progression reduces the
mography (PET) [5], single photon emission computed NMDA receptor affinity, it contributes to thesafety and
tomography (SPECT) [6,7], functional magnetic reso- efficacy of memantine as a neuroprotective agent. How-
nance imaging (fMRI) and, more recently, magnetic re- ever, the process also makes memantine less effective
sonance diffusion and diffusion tensor spectroscopy to than high-affinity antagonists (for example, ketamine)
identify the brain structures activated during episodes of in the treatment of chronic pain [21,26]. Nevertheless,
pain in patients and controls [8]. These structures recent research has highlighted the efficacy of meman-
include the primary and secondary sensory and motor tine for the treatment of complex regional pain syn-
cortices, insula, anterior cingulate cortex, thalamus, drome[27]andphantomlimbpain[28],whichsuggests
dorsolateral prefrontal cortex and basal ganglia - regions that the extent of analgesia depends on the type of pain
that have been named the ‘pain matrix’ because they are being treated. We have carried out a preliminary open,
activated in response to a painful stimulus. A growing uncontrolled, three-month follow-up study of meman-
body of evidence suggests that glutamate (Glu), an exci- tine in patients with FM (EudraCT Number: 2011-
tatory neurotransmitter in the central nervous system, 000802-23). As will be commented later on in the
may play a part in the pathophysiology of FM, given that discussion, in this preliminary study there was a trend
its concentration is elevated in the insula [8], hippocam- toward improvement in pain, but it was not significant,
pus[9]andposterior cingulate cortex [10,11]. probably due to the small sample size. Glutamate levels
Asaconsequence, anumberofauthorshavesuggested did not show modifications from baseline as a lon-
that glutamate-blocking drugs may be useful in the ger period follow-up is usually required. After three
treatment of FM [12]. Studies suggest that memantine months, memantine-treated patients in our study
could reduce the harmful effects that result from the ex- showed significant improvement in cognitive function,
cessively high levels of brain glutamate found in a num- symptoms of depression and global functioning (per-
ber of conditions, such as FM [13]. Memantine is not sonalcommunication).
believed to act by reducing levels of glutamate or by
preventing its release; rather, it is believed to reduce Aims
glutamate’sneurotoxiceffectbyblockingtheN-methyl-D- The primary objective is to evaluate the efficacy of
aspartate (NMDA) receptor, thereby preventing the entry memantine in the treatment of pain (pain perception) in
of excess calcium [14]. Memantine belongs to the family patientswith fibromyalgia.
ofdrugsknownasNMDAreceptorantagonists. Thesecondaryobjectivesofthisstudyarethefollowing:
NMDA receptor antagonists have neuroprotective and 1)Toevaluatetheefficacyofmemantineinthetreatment
analgesic properties and are widely used in clinical prac- ofothersymptomsoffibromyalgia,suchaspainthreshold,
tice. Dextromethorphan and ketamine have demon- impaired cognitive function, reduced health status, an-
strated efficacy in the treatment of pain in fibromyalgia xiety,depression,qualityoflifeandperceivedimprovement
[15,16], although their use as long-term treatment of symptoms; 2) To evaluate the efficacy of memantine in
Olivan-Blázquezetal.Trials2013,14:3 Page3of9
http://www.trialsjournal.com/content/14/1/3
reducing brain levels of glutamate, as measured by spec- period (including a dose adjustment period of one
trometry,inpatientswithfibromyalgia. month).ThestudyflowchartisshowninFigure1.
Settingandstudysample(Participants)
Methods PatientsdiagnosedwithFMwillberecruitedforinclusion
Studydesign inthestudyfromprimaryhealthcarecentersinZaragoza,
Doubled-blind, multicenter, parallel randomized clinical Spain, upon fulfilment of the following selection criteria:
trialwith six-monthfollow-up. a)subject(malesandfemales)withagebetween18and65
Seventy patients with FM will be recruited for inclu- years;b)abilitytounderstandSpanish;c)diagnosisofFM
sion in the study upon fulfilment of selection criteria. performed by a rheumatology specialist according to the
The patients will be randomized in two parallel groups: American College of Rheumatology (ACR 1990) diagnos-
a treatment group (n = 35), which will be given 20 mg tic criteria; d) signing of an informed consent form; and
of memantine daily; a control group (n = 35), which will e)use of birthcontrol duringthe study in the case of fer-
receive a placebo. There will be a six-month follow-up tilewomen.
TIMELINE
60 patients from primary care that
fulfil selection criteria.
Recruitment Signing of informed consent
Randomisation
Memantine Group (n=30) Placebo Group (n=30)
1stvisit (baseline assessment). Evaluation of study
Initial variables (neuroimaging tests and administration
assessment of questionnaires)
Administration of memantine/placebo
Progressive increase of the dose to a daily intake of
20 mg of memantine/placebo.
1st month 2ndvisit (1 month assessment): administration of
questionnaires.
Dose of 20mg of memantine/placebo
3rdvisit (3 month assesment). Evaluation of study
3rd month
variables (administration of questionnaires)
Dose of 20mg of memantine/placebo.
4thvisit (6 month assessment). Evaluation of
6th month
variables (neuroimaging tests and questionnaires)
Figure1Flowchartofthestudy.
Olivan-Blázquezetal.Trials2013,14:3 Page4of9
http://www.trialsjournal.com/content/14/1/3
Patients must not fulfil any of the following criteria at A patient is to withdraw from the trial if he or she
the time of inclusion: a) be undergoing drug treatment withdraws informed consent, if the researcher feels that
for fibromyalgia. In these cases, the patients will discon- he or she should withdraw from the study for reasons of
tinue treatment and go through a washout period of one safety/efficacy or in the best interest of the patient, or if
week to minimize the influence of the medication on the patient does not comply with the treatment for more
brain imaging. During that week, the patient may take, if than sevenconsecutive days.
necessary, low doses of analgesics such as tramadol or
paracetamol; b) be taking memantine, or have taken
Intervention
memantine during the one year prior to recruitment; c)
Treatmentgroup
be suffering from an Axis I psychiatric disorder, as The treatment group will receive the study drug,
diagnosed using the Structured Clinical Interview for
memantine, in a dose of 20 mg daily for six months, in-
DSM-IV (SCID-I), that might hinder adherence to the
cludingaone-month titration period.
protocol (for example, dementia, alcohol and/or sub-
stance abuse/dependence, or schizophrenia); d) be preg-
Controlgroup
nant or breast-feeding; e) have a hypersensitivity to the
The control group will receive a daily dose of placebo
active ingredient memantine, or to the excipients; f)
(coated pills with the same external appearance as the
have conditions that require special precautions when
active drug) and will take the same number of pills as
administering memantine according to the summary of
the treatment group. Administration will be through the
product characteristics (namely, epilepsy, circumstances
oral route.
that may cause high urinary pH such as Proteus urinary
This is a double-blind study. Consequently, the
infection,renaltubularacidosisor vegetariandiet,recent
patientswillberandomized,andneitherthepatient,the
myocardial infarction, congestive heart disease and un-
doctor nor the researcher administering questionnaires
controlled arterial hypertension); g) have clinically sig-
or spectrometry will know to which group the patient
nificant and active evidence of liver or kidney disease,
has been assigned. The recommended dose of meman
hematological, respiratory, endocrine or cardiovascular
tine in adults is 20 mg once a day. To minimize adverse
disease or disorders. However, patients with controlled
effects, 20 mg doses will be reached by the following ti-
diabetes, controlled hypertension and complete or in-
tration schema: first week, 5 mg daily; second week, 10
complete right bundle branch block can be included in
mg daily; third week, 15 mg daily; fourth week, 20 mg
the study; h) use drugs that may cause relevant
daily.
interactions with memantine according to the summary
Thenumber oftabletsineachdispensed container will
of product characteristics, namely NMDA receptor
be monitored at each evaluation, keeping track of the
antagonists (for example, amantadine, ketamine, dex-
number the patient should have taken and how many
tromethorphan), L-Dopa, dopamine agonists and cholin-
should remain on completion of the treatment. The
ergic agonists; or i) use non-permitted concomitant
route of administration is oral, in the form of film-
medication during the week prior to the first evaluation
coated tablets. The drug for use in the study (both
visit or be expected to require treatment with at least
memantine pills and placebo) will be prepared, condi-
oneofthedrugs notpermittedduringthestudy, namely,
tioned and released by one qualified person according to
antidepressants (for example, duloxetine, venlafaxine,
the principles of Good Manufacturing Practice, under
mirtazapine, bupropion, SSRIs, and so on), analgesics
the responsibility of H Lundbeck A/S. Upon completion
(for example, pregabalin, gabapentin, opiates, and so on)
of the trial, patients will continue with standard FM
or other drugs. During this week, patients may take
treatmentaccordingtoclinicalpracticeguidelines.
analgesics such as tramadol or paracetamol if needed,
but only sporadically to minimize the influence of the
medicationonbrain images. Outcomesandmeasurements
There is currently no approved treatment with an in- Mainoutcomevariables
dication for the treatment of FM. In 2007, the FDA The main efficacy variable is the improvement in the
approved pregbalin as the first drug for the treatment treatment of pain, specifically pain perception. Perceived
of FM symptoms in the USA. This agency has sub- pain will be measured by means of a visual analogue
sequently approved duloxetine and milnacipran for scale (VAS) from 0 to 100. The psychometric usefulness
the same indication. Although these drugs are com- ofVAShasbeenwidelydemonstrated [30].
mercially available in Europe with other indications,
European regulatory authorities have recently refused Secondaryvariables
to widen their approval to include them for the treat- Secondary efficacy variables are modifications made to
mentofFM[29]. values of the following clinical variables: pain threshold,
Olivan-Blázquezetal.Trials2013,14:3 Page5of9
http://www.trialsjournal.com/content/14/1/3
cognitivestate,health status,stateofanxietyanddepres- cingulate cortex, as assessed by magnetic resonance
sion,clinicalimprovement impression andquality oflife. spectroscopy(MRS).
Pain threshold will be measured by means of a sphyg- The following parameters will used for Magnetic
momanometer, a widely used clinical test that has been Resonance Spectroscopy (MRS): a T2-weighted coronal
demonstrated to be useful for identifying FM patients image with repetition time (TR) = 5,350 ms, echo delay
[31]. It is recommended that the blood pressure cuff time (ET) = 85 ms, 90° flip angle, number of excitations =
should be inflated in increments of approximately 10 2, imaging matrix = 320 × 256, field of view (FOV) = 24
mmHg up to 180 mmHg or to the point that pain cm × 24 cm and slice thickness/distance = 5/0 mm and
appears. Healthy persons tend to feel pain when the takenontheplanethroughtheinternalauditorycanaland
pressure cuff is inflated to 160 mmHg or more, while cerebralpedunclewillbeusedtolocatevolumesofinterest
FM patients generally present pain at pressures between (VOIs)(2×2×2cm)inbothhippocampi.AT1-weighted
100 and 110 mmHg, or even lower. Blood pressure medial sagittal image (TR = 560 ms, ET = 12 ms, 90° flip
should be recorded to adjust for the effect of hyperten- angle, number of excitations = 1, imaging matrix = 256 ×
siononpainthreshold. 160,FOV=24cm ×24cm,slicethickness/distance=5/0
The rest of the secondary efficacy variables will be mm)willbeobtainedtolocateavoxelintheposteriorcin-
measured by structured exams. The Cognition Mini- gulatecortex,andaparasagittalT1-weightedimagewillbe
Exam (MEC) will measure cognitive state. The MEC is a taken with respect to the right of the corpus callosum
structured scale that consists of 35 points grouped into planetolocatevoxelsontheanteriorandposteriorinsular
seven categories: orientation to place, orientation to regions. Proton magnetic resonance spectroscopy will be
time, recall, attention and concentration, memory, lan- performedwithashortETof35ms,aTRof2,000msand
guage and visual construction. In non-geriatric popu- 128 accumulations by means of thesingle-voxel spin-echo
lations (under the age of 65) such as the sample for this technique, which uses selective excitation with gradient
study, the threshold that suggests a‘likely case’ of a cog- spoiling for water suppression. Spectral acquisition will
nitive disorder is 27 points and lower. This test is the be accomplished by means of the PROBE/PRESS (pro-
validated Spanish-language version of the Mini-Mental ton brain spectroscopy/point-resolved spatially localized
State Examination(MMSE)[32]. spectroscopy) technique. LCModel (version 6.0) user-
Health status in FM will be measured by means of the independent frequency domain-fitting software will be
Fibromyalgia Impact Questionnaire (FIQ). The FIQ is a used to quantify absolute concentrations of brain meta-
ten-item self-assessment questionnaire that measures the bolites in mmol/kg. Eddy current correction will be ap-
health status of FM patients. The validated Spanish- plied, and the use of the internal water signal will be the
language version of this questionnaire will be used [33]. reference from which the absolute metabolite con-
Anxiety and depression will be measured by means of the centrations will be calculated.Inadditiontothe individual
Hospital Anxiety Depression Scale (HADS). HADS is a analysisofglutamate(Glu),creatine(Cr),N-acetylaspartate
scale based on self-report that was developed to detect the (NAA) and myo-inositol (mI) compounds, the aggregate
presence of depression and anxiety disorders in medical concentrations of the following three compound pairs will
patientsinprimarycaresettings.Itcontainsfourteenitems be studied: NAA + N-acetyl-aspartyl-glutamate (NAA +
scored on a four-point Likert-type scale. This scale is NAAG), referred to as the total phosphocholine (PCh);
comprised of two subscales that separately assess depres- Glycerophosphorylcholine (GPC), referred to as the total
sionandanxiety.ThevalidatedSpanish-languageversionof Cho;andglutamate+glutamine,referredtoasGlx.Abso-
thisscalewillbeused[34].Qualityoflifewillbemeasured lute metabolite concentrations will be considered only if
by means of the EuroQol 5D (EQ5D) questionnaire. This theCramér-Raolowerboundsarelowerthan20%,indicat-
questionnaireisastandardizedinstrumentusedasameas- ing that these metabolite concentrations can be reliably
ure of health outcomes. It is applied to a wide range of calculated.Chemicalconcentrationscanbeextractedauto-
healthconditionsandtreatmentsandprovidesasimplede- matically from MR spectra with well-documented time
scriptive profile and a single index value for health status. domains,andthespectralfrequencycanbefittedwithsoft-
The Spanish-language version of this questionnaire will be ware packages such as LCModel (Stephen Provencher,
used [35]. Perceived clinical improvement will be assessed Oakville,Ontario,Canada).
bytheClinicalGlobalImpressionscale(CGI). Table 1 shows the procedure and evaluation times of
thetrial.
Magneticresonancespectroscopy Samplesize
Another relevant secondary efficacy variable in this The choice was made to have a sample comprising 70
studyisthelevelofglutamatein differentregionsofthe subjects (n = 35 subjects in each of the two treatment
brain, namely the insula, hippocampus and posterior groups) because this sample size will enable us to obtain
Olivan-Blázquezetal.Trials2013,14:3 Page6of9
http://www.trialsjournal.com/content/14/1/3
Table1Procedureandevaluationtimesofthetrial
Visit Baseline 1month 3months 6months
Visitnumber 1 2 3 4
Recruitment&Clinicalassessment
Signingofinformedconsent X
Diagnosis X
Inclusion/exclusioncriteria X
Sociodemographicvariables X
Painassessedwithsphygmomanometer X X X X
PainassessedwithAnalogueVisualScale X X X X
Cognitivefunction(assessedwithMEC) X X X X
Globalfunction(evaluatedbyFIQ) X X X X
Anxietyanddepression(assessedwithHADS) X X X X
Qualityoflife(assessedbyEuroQol5D) X X X X
ClinicalGlobalImpression(CGI) X X X X
Neuroimageassessment
Magneticresonancespectroscopy X X
Safetyassessments
Adverseevents X X X
Otherstudyprocedures
Deliveryoftheresearchdrug X X X
Inventoryofdrugtakenbackandused X X X
Concomitantpharmacologicaltreatment X X X X
reliable results and to ensure thatboth treatment groups The assignment will be carried out by an independent
are of equal sizes at the start of the study. Smaller sam- person belonging to REDIAPP (Research Network on
ple sizes have been used in the identification of signifi- Preventative Activities and Health Promotion) who is
cant differences in glutamate levels in different brain not involved in the study. The random allocation se-
regions betweenFMpatientsandcontrols. quence will be implemented by telephone. The sequence
Using the results of the pilot study and taking reduc- will be concealed until interventions are assigned.
tion of pain (as determined using the Pain Visual Patients agree to participate before the random alloca-
Analogue Scale and sphygmomanometry) as the primary tion and without knowing to which treatment they will
variables, a pre-intervention mean score of 56 (SD 14.9) be assigned. According to the intention-to-treat (ITT)
was obtained with the VAS, and 104 (SD 30.8) with principle, any patients randomized and allocated to any
sphygmomanometry. After treatment, the means dec- arm that starts the study will be included in the analysis.
reased to 44 (SD 16.9) and 85 (SD 20.6), respectively. Pharmacological treatment will be administered by one
Therefore, assuming a 95% confidence interval and a psychiatrist (JGC). Study personnel conducting psycho-
power of 80%, we find that we require a sample size of logical assessments (BO, MCPY, MP, BM) will be
28 individuals in each group for the VAS finding and of maskedtotheparticipants'treatment.
30 individuals in each group for the reduction measured
by sphygmomanometry. The resulting sample size that Statisticalmethods
will enable us to analyse the final variable will be 60 Analysisstrategy
individuals, with 30 being assigned to each of the two Theanalysisofclinicalefficacywillmakeuseofintention-
groups.Basedonpreviousstudies(personalcommunica- to-treat analysis. The Last-Observation-Carried-Forward
tion), an attrition rate of 10% can be expected. There- (LOCF) method will be used for handling missing data.
fore, the final number of recruited patients will be 35 An initial comparison will be made between both groups,
patientsfor eacharm. examining key variables to establish the groups’ baseline
comparability after randomization. To describe the quan-
Randomization,allocationandmaskingofstudygroups titative variables, means and standard deviations will be
Each patient will be assigned to one of the two groups calculated when they fulfil normality criteria. The Chi-
using a computer-generated random number sequence. squaredtestwillbeusedforqualitativevariablessuchasa
Olivan-Blázquezetal.Trials2013,14:3 Page7of9
http://www.trialsjournal.com/content/14/1/3
number of socio-demographic ones. The differences be- pharmacological intervention. They will also be in-
tween clinical variables at baseline, one month, three formed that they will be participating voluntarily and
monthsandsixmonthswillbecalculatedusinganalysisof that they can choose to withdraw at any time with the
covariance(ANCOVA)adjustedbybaseline. guarantee that they will continue to receive the treat-
To study the main variable, analysis of variance will be mentconsideredmostappropriate bytheirdoctor.
performed on repeated measurements, including all the With regard to the potential risks of the study, data
evaluations over time. The main variable (pain percep- gathering involves no risks to the subjects participating
tion) will be taken to be a continuous variable for this in the study, and the neuroimaging techniques that will
purpose. The models will include adjustments for the be performed are non-invasive techniques that do not
baseline pain value and for any other variable that may place subjects in any danger. With regard to the poten-
have shown differences in the baseline measurement. tial benefits, if the hypothesis of the study, the patients
Possible group × time interactions will be studied by in the treatment group will be able to control their pain
means of Mixed Factor ANOVA. Additionally, linear re- during the course of this research. The results will be
gression models will be used to compare the differences communicated to the scientific community and to the
between the two groups for each of the evaluations over relevant institutions, which will lead to the expected
time as compared to baseline. Similar analyses will be benefits being obtained for society and for people
performedonthe othersecondaryclinicalvariables. diagnosed with FM. Nevertheless, there is a possibility
To analyze the differences found by the neuroimaging that there will be no benefit gained from participating in
test (spectrometry) taken at baseline and at six months thisstudy.
follow-up, Student’s t-test for paired measurements will The study followsHelsinki Convention norms withpos-
be calculated. The relation between neuroimaging terior modifications, and the Declaration of Madrid of the
variables and clinical variables will be assessed by means World Psychiatric Association. The Study Protocol was
ofSpearman’sRhonon-parametric test. approved by the Clinical Research Ethics Committee of
In case of lost values, a sensitivity analysis will be Aragón (June 2012) and the Medicines and Health
made to estimate the effect of the lost values on the ProductsAgencyofSpain(EUDRACT2011-006244-73).
results. These values will also be replaced using several
approaches, such as last recorded values and impu- Discussion
tations. Statistical analyses will be performed with the This is the first randomized, controlled study of
Statistical Package for the Social Sciences (SPSS) 19.0 memantine for the treatment of fibromyalgia. Memantine
statistical software package (SPSS Inc., 233 South might be expected to be useful for the treatment of FM
Wacker Drive, 11th Floor, Chicago, IL 60606–6412), based on its pharmacological effects, as described in the
with P-values below 0.05consideredsignificant. introduction [13,14]. In addition, our research group have
recently carried out an exploratory, uncontrolled study
Safetyandmonitoring witha smallsample(n=10)andathree-monthfollow-up
Any adverse events and serious adverse events will be that found significant improvement of global function (as
recorded and a determination will be made as to measuredbyFIQ),depression(asevaluatedbyHADS)and
whether any medical action required is related with the cognitive function (as assessed with MEC) in the study
administered drug. Any adverse event related to the patients at three-month follow-up (personal communi-
studydrugwillberecordedandmonitoreduntilitsreso- cation, Study EudraCT Number: 2011-000802-23).
lution or stabilization, and the regulatory notification However, at three-month follow-up, glutamate levels
processwillbefollowedinaccordancewiththepertinent did not decrease in any of the brain areas studied (per-
legislation in force (timeframes, unmasking, and so on). sonalcommunication).
Study participants will be requested to report to the Based on previous preliminary pilot studies (personal
researchers any adverse effect (serious or otherwise) that communication) a significant improvement in key clin-
may arise in the period between visits or any other ical symptoms in fibromyalgia, such as depression, cog-
circumstances that may lead to their withdrawal from nition and global function can be expected. Cognition is
thestudy. one of the most impairing symptoms in fibromyalgia
[36] and is strongly correlated with depression [37].
Ethicalaspects Cognition in these patients seems to be related to brain
Informed consent will be obtained from the participants morphology changes that could be improved with
before they are aware of which group they are to be memantine [38]. Depression, another frequent and quite
included in. Before they give their consent, the patients disabling symptom in fibromyalgia [39] is also expected
will be provided with a general overview of the aims and to improve with memantine due to its pharmacological
characteristics of the study and the psychological and effect[40].
Olivan-Blázquezetal.Trials2013,14:3 Page8of9
http://www.trialsjournal.com/content/14/1/3
Pain,themostrelevantsymptominfibromyalgiaandthe Expectedfinalizationofpatientrecruitment:31December
main outcome of our study, was not significantly relieved 2012.
inouropen,uncontrolledstudy(personalcommunication), Expected finalization of patient monitoring period: 30
probably due to its small sample size. However, previous June2013.
pilot studies (26 to 28) suggest memantine is effective in Publicationofresults:December 2013.
relievingpainingeneral.Inaddition,spectrometricstudies
Abbreviations
in fibromyalgia found a strong correlation between high
ACR:AmericanCollegeofRheumatology;ANOVA:Analysisofvariance;
levelsofglutamateanddepressionandglobalfunction,but CGI:ClinicalGlobalImpression;CI:Confidenceinterval;Cr:Creatine;
alsowithpain(10). EQ5D:EuroQol5D;ET:Echodelaytime;FIQ:FibromyalgiaImpact
Questionnaire;FM:Fibromyalgia;fMRI:Functionalmagneticresonance
Treatment with memantine decreases the Glu/Cr
imaging;FOV:Fieldofview;Glu:Glutamate;GPC:Glycerophosphorylcholine;
(glutamate/creatine) ratio inthe lefthippocampal region, HADS:HospitalAnxietyDepressionScale;MEC:CognitionMini-Exam;
which may account for its anti-excitotoxic property [41]. LOCF:Last-Observation-Carried-Forward;mI:myo-inositol;MMSE:Mini-Mental
StateExamination;MRS:Magneticresonancespectroscopy;NAA:
Other studies have observed an increase in cingulate
N-acetylaspartate;NAAG:N-acetyl-aspartyl-glutamate;NMDA:
glutamatergic turnover [42]. There is evidence from MRS N-methyl-D-aspartate;PCh:Phosphocholine;PET:Positronemission
studies of other excitotoxicity-related conditions, such as tomography;PROBE/PRESS:Protonbrainspectroscopy/point-resolved
amyotrophic lateral sclerosis [43] and Huntington’s di- spatiallylocalizedspectroscopy;REDIAPP:ResearchNetworkonPreventative
ActivitiesandHealthPromotion;SD:Standarddeviation;SCID-I:Structured
sease [44], that glutamate is elevated in several brain ClinicalInterviewforDSM-IVaxis-I;SPECT:Singlephotonemissioncomputed
regions in these disorders. Thus, the decrease in tomography;TR:Repetitiontime;VAS:Visualanaloguescale;VOIs:Volumes
ofinterest.
hippocampalglutamatewithmemantine(withoutchanges
in NAA/Cr) could explain some of memantine’s thera- Competinginterests
peuticeffects. Theauthorsdeclarethattheyhavenoconflictsofinterest.Theresearch
groupthatdesignedanddevelopedthisstudyisfinancedbythe
In this study, there is no clear hypothesis as to
DepartmentofScience,TechnologyandUniversityoftheGovernmentof
whether memantine is able to reduce glutamate levels in AragonandbytheCarlosIIIInstituteofHealth,whichisattachedtothe
the affected brain areas, or, independent of this possibil- SpanishMinistryofScienceandInnovation.
ity, whether the drug will be an effective treatment for
Authors´contributions
different clinical symptoms of fibromyalgia, particularly
JGC,RM,BO,MR,YLdHandASBaretheprincipalresearchersanddeveloped
in the cognitive domain. It is possible that it could re- theoriginalideaforthestudy.Thestudydesignwasfurtherdevelopedby
duce glutamate without ameliorating any symptoms. Al- MCPY,MPandBM.EAwilldevelopthestatisticalmethods.Allauthorshave
readandcorrecteddraftversionsandapprovedthefinalversion.
ternatively, it may not reduce glutamate while improving
some target symptoms. In any case, new hypotheses as Acknowledgements
to the aetiology of fibromyalgia, and of the origins of the ThestudyhasbeenfundedbyagrantfromtheMinistryofHealthofthe
GovernmentofSpain(EC11-387).WethankReddeInvestigaciónen
symptoms of the disease may be developed based on the
ActividadesdePrevenciónyPromocióndelaSalud(ResearchNetworkon
resultsofthispreliminaryresearch. PreventativeActivitiesandHealthPromotion)(REDIAPP-GRD06/0018/0020),
The main strength of the study is that this is the first NododeAragón,foritssupportinthedevelopmentofthisstudy.Weare
gratefulforthesupportofH.LundbeckA/Sinpreparingthemedication,
randomized controlled trial of memantine for the treat-
bothactiveandplacebo.
ment of FM. The main limitations specific to the spec-
trometric study will be the following: 1) The proportion Authordetails
1DepartmentofPsychologyandSociology,UniversityofZaragoza,Zaragoza,
of grey matter/white matter and CSF within the Spain.2ReddeActividadesPreventivasydePromocióndelaSalud
corresponding voxel will not be calculated, so the 1H- (REDIAPP)(RD06/0018),InstitutoAragonésdeCienciasdelaSalud(IACS),
MRS signals, which are produced by both grey and Aragón,Spain.3UnidadEpidemiologíaClínica,Hospital12deOctubre,CIBER
EpidemiologíaySaludPública,Madrid,Spain.4CentrodeSaludArrabal,
white matter, will represent an amalgamation of mul- Zaragoza,Spain.5InstitutUniversitarid'InvestigacióenCiènciesdelaSalut
tiplecelltypes.Differencesinthefractionsofeachvoxel (IUNICS),UniversityofBalearicIslands,PalmadeMallorca,Spain.6Serviciode
consisting of different tissue types may result from ana- Psiquiatría,HospitalMiguelServetyUniversidaddeZaragoza,Zaragoza,
Spain.7DepartmentofPsychiatry,MiguelServetHosìal,AvdaIsabelLa
tomicaldifferencesassociatedwiththeconditionsunder
Catolica1,Zaragoza50.009,Spain.
investigation. 2) Metabolites will not be quantified by
segmentation and CSF-correction, so the measured Received:7August2012Accepted:18December2012
Published:3January2013
levelsofglutamatewilloftenbecontaminatedinpartby
glutamine. Furthermore, glutamate exists in the neuro- References
transmitterpoolaswellasinthemetabolicpool. 1. WolfeF,SmytheHA,YunusMB,BennetRM,BombardierC,GoldenbergADL:
AmericanCollegeofRheumatology1990.Criteriafortheclassificationof
fibromyalgia.ReportoftheMulticenterCriteriaCommittee.ArthrRheum
Trial status 1990,33:160–172.
Inclusionofpatientscommenced15September 2012. 2. BrancoJC,BannwarthB,FaildeI,AbelloCarbonellJ,BlotmanF,SpaethM,
SaraivaF,NacciF,ThomasE,CaubèreJP,LeLayK,TaiebC,Matucci-CerinicM:
Initialrecruitmentofpatientscommenced15September
Prevalenceoffibromyalgia:asurveyinfiveEuropeancountries.
2012. SeminArthritisRheum2010,39:448–453.
Olivan-Blázquezetal.Trials2013,14:3 Page9of9
http://www.trialsjournal.com/content/14/1/3
3. GameroRuizF,GabrielSánchezR,CarbonellAbellóJ,TorneroMolinaJ, amputationorsurgery:arandomized,double-blinded,cross-overstudy.
Sanchez-MagroI:PaininSpanishrheumatologyoutpatientoffices: AnesthAnalg2000,91:960–966.
EPIDORepidemiologicalstudy.RevClinEsp2005,205:157–163. 27. SinisN,BirbaumerN,GustinS,SchwarzA,BredangerS,BeckerST,UnertlK,
4. GarciaCampayoJ,MagdalenaJ,FernándezE,MagallónR,SalasM,SobradielN: SchallerHE,HaerleM:Memantinetreatmentofcomplexregionalpain
Effectivenessoftreatmentsforfibromyalgiadependingoflevelofcare:a syndrome:apreliminaryreportofsixcases.ClinJPain2007,23:237–243.
meta-analysis.ArthitisResTher2008,10:R81. 28. HackworthRJ,TokarzKA,FowlerIM,WallaceSC,Stedje-LarsenET:Profound
5. MountzJM,BradleyLA,ModellJG,AlexanderRW,Triana-AlexanderM, painreductionafterinductionofmemantinetreatmentintwopatients
AaronLA,StewartKE,AlarconGS,MountzJD:Fibromyalgiainwomen. withseverephantomlimbpain.AnesthAnalg2008,107:1377–1379.
Abnormalitiesofregionalcerebralbloodflowinthethalamusare 29. StaudR:Drugstotreatfibromyalgia-thetransatlanticdifference.
associatedwithlowpainthresholdlevels.ArthrRheum1995,38:926–938. CurrOpinInvestigDrugs2010,11:16–18.
6. KwiatekR,BarndenL,TedmanR,JarretR,ChewJ,RoweC,PileK:Regional 30. VargasA,VargasA,Hernández-PazR,Sánchez-HuertaJM,Romero-RamírezR,
cerebralbloodflowinfibromyalgia:single-photon-emissioncomputed Amezcua-GuerraL,KoohM,NavaA,PinedaC,Rodríguez-LealG,Martínez-LavínM:
tomographyevidenceofreductioninthethalami.ArthrRheum2000, Sphygmomanometry-evokedallodynia–asimplebedsidetestindicativeof
43:2823–2833. fibromyalgia:amulticenterdevelopmentalstudy.JClinRheumatol2006,
7. García-CampayoJ,Sanz-CarrilloC,BaringoT,CeballosC:SPECTscanin 12:272–274.
somatizationdisorderpatients.Australia&NZJPsychiatry2001,35:359–363. 31. SriwatanakulK,KelvieW,LasagnaL:Studieswithdifferenttypesofvisual
8. HarrisRE,SundgrenPC,CraigAD,KirshenbaumE,SenA,NapadowV,ClauwDJ: analogscalesformeasurementofpain.ClinPhrmacolTher1983,
Elevatedinsularglutamateinfibromyalgiaisassociatedwithexperimental 34:234–239.
pain.ArthritisRheum2009,60:3146–3152. 32. LoboA,SazP,MarcosG,DíaJL,delaCámaraC,VenturaT,MoralesAsínF,
9. ValdésM,ColladoA,BargallóN,VázquezM,RamiL,GómezE,SalameroM: FernandoPascualL,MontañésJA,AznarS:Revalidationand
Increasedglutamate/glutaminecompoundsinthebrainsofpatients standardizationofthecognitionmini-exam(firstSpanishversionofthe
withfibromyalgia:amagneticresonancespectroscopystudy. Mini-MentalStatusExamination)inthegeneralgeriatricpopulation.
ArthritisRheum2010,62:1829–1836. MedClin(Barc)1999,112:767–774.
10. FayedN,Garcia-CampayoJ,MagallónR,Andrés-BergarecheH,LucianoJV, 33. RiveraJ,GonzalezT:TheFibromyalgiaImpactQuestionnaire:avalidated
AndresE,BeltránJ:Localized1H-NMRspectroscopyinpatientswith Spanishversiontoassessthehealthstatusinwomenwithfibromyalgia.
fibromyalgia:acontrolledstudyofchangesincerebralglutamate/glutamine, ClinExpRheumatol2004,22:554–560.
inositol,choline,andN-acetylaspartate.ArthritisResTher2010,12:R134. 34. TejeroA,GuimeráEM,FarréJM,PeriJM:UsoclíınicodelHAD(Hospital
11. HarrisRE:Elevatedexcitatoryneurotransmitterlevelsinthefibromyalgia AnxietyandDepressionScale)enpoblaciónpsiquiátrica:unestudiode
brain.ArthritisResTher2010,12:141. susensibilidad,fiabilidadyvalidez.RevDepPsiquiatrFacMedBarc1986,
12. ChoiDW,KohJY,PetresS:Pharmacologyofglutamateneurotoxicityin 13:233–238.
corticalcellculture:attenuationbyNMDAantagonists.JNeuroscience 35. BadíaX,RosetM,HerdmanMandSeguraA:Laversiónespañoladel
1988,8:185–196. EuroQol:descripciónyaplicaciones.MedClin(Barc)1999,
13. JohnsonJW,KotermanskiSE:Mechanismofactionofmemantine. 112(Suppl1):79–86.
CurrOpinPharmacology2006,6:61–67. 36. GlassJM:Fibromyalgiaandcognition.JClinPsychiatry2008,
69(Suppl2):20–24.
14. StaudR,VierckCJ,RobinsonME,PriceDD:EffectsoftheN-methyl-D-
37. SuhrJA:Neuropsychologicalimpairmentinfibromyalgia:relationto
aspartatereceptorantagonistdextromethorphanontemporal
depression,fatigue,andpain.JPsychosomRes2003,55:321–329.
summationofpainaresimilarinfibromyalgiapatientsandnormal
controlsubjects.JPain2005,6:323–332. 38. LuerdingR,WeigandT,BogdahnU,Schmidt-WilckeT:Workingmemory
performanceiscorrelatedwithlocalbrainmorphologyinthemedial
15. WeinbroumAA,RudickV,ParetG,Ben-AbrahamR:Theroleof
dextromethorphaninpaincontrol.CanJAnaesth2000,47:585–596. frontalandanteriorcingulatecortexinfibromyalgiapatients:structural
correlatesofpain-cognitioninteraction.Brain2008,131(12):3222–3231.
16. Graven-NielsenT,Aspegren-KendallS,HenrikssonKG,BengtssonM,
39. AgugliaA,SalviV,MainaG,RossettoI,AgugliaE:Fibromyalgiasyndrome
SörensenJ,JohnsonA,GerdleB,Arendt-NielsenL:Ketaminereduces
anddepressivesymptoms:comorbidityandclinicalcorrelates.JAffect
musclepain,temporalsummation,andreferredpaininfibromyalgia
patients.Pain2000,85:483–491. Disord2011,128:262–266.
40. OwenRT:Glutamatergicapproachesinmajordepressivedisorder:focuson
17. AlviarMJ,HaleT,DungcaM:Pharmacologicinterventionsfortreating
ketamine,memantineandriluzole.DrugsToday(Barc)2012,48:469–478.
phantomlimbpain.CochraneDatabaseSystRev2011,7(12):CD006380.
41. GlodzikL,KingKG,GonenO,LiuS,DeSantiS,deLeonMJ:Memantine
18. ChenHSV,LiptonSA:ThechemicalbiologyofclinicallytoleratedNMDA
receptorantagonists.JNeurochem2006,97:1611–1626. decreaseshippocampalglutamatelevels:amagneticresonance
spectroscopystudy.ProgNeuropsychopharmacolBiolPsychiatry2008,
19. ReisbergB,DoodyR,StöfflerA,SchmittF,FerrisS,MöbiusHJ:Memantine 32:1005–1012.
inmoderate-to-severeAlzheimer’sdisease.NEnglJMed2003,
42. RowlandLM,BustilloJR,MullinsPG,JungRE,LenrootR,LandgrafE,BarrowR,
348:1333–1341.
YeoR,LaurielloJ,BrooksWM:Effectsofketamineonanteriorcingulate
20. TariotPN,FarlowMR,GrossbergGT,GrahamSM,McDonaldS,GergelIand
glutamatemetabolisminhealthyhumans:a4-TprotonMRSstudy.
MemantineStudyGroup:Memantinetreatmentinpatientswith AmJPsychiatry2005,162:394–396.
moderatetosevereAlzheimer’sdiseasealreadyreceivingdonepezil:a
43. PioroEP,MajorsA,MitsumotoH,NelsonDR,NgT:1H-MRSevidenceof
randomizedcontrolledtrial.JAMA2004,291:317–324.
neurodegenerationandexcessglutamate+glutamineinALSmedulla.
21. ReisbergB,DoodyR,StofflerA,SchmittF,FerrisS,MöbiusHJ:A24-week Neurol1999,53:71–79.
open-labelextensionofmemantineinmoderatetosevereAlzheimer’s
44. ReynoldsNC,ProstRW,MarkLP:Heterogeneityin1H-MRSprofilesof
disease.ArchNeurol2006,63:49–54.
presymptomaticandearlymanifestHuntington'sdisease.BrainRes2005,
22. ParkJW,SuhGI,ShinHE,ParkGE:Influenceofmemantineonnociceptive 1031:82–89.
responsesofthetrigeminocervicalcomplexafterformalininjection.
Cephalalgia2012,32(4):308–316.
doi:10.1186/1745-6215-14-3
23. SuzukiM:RoleofN-methyl-D-aspartatereceptorantagonistsinpostoperative
Citethisarticleas:Olivan-Blázquezetal.:Evaluationoftheefficacyof
painmanagement.CurrOpinAnaesthesiol2009,22(5):618–622.
memantineinthetreatmentoffibromyalgia:studyprotocolfora
24. BigalM,RapoportA,SheftellF,TepperD,TepperS:Memantineinthe doubled-blindrandomizedcontrolledtrialwithsix-monthfollow-up.
preventivetreatmentofrefractorymigraine.Headache2008,48(9):1337–1342. Trials201314:3.
25. ReclaJM,SarantopoulosCD:Combineduseofpregabalinandmemantine
infibromyalgiasyndrometreatment:anovelanalgesicand
neuroprotectivestrategy?MedHypotheses2009,73(2):177–183.
26. NikolajsenL,GottrupH,KristensenAGD,JensenTS:Memantine(aN-methyl-D-
aspartatereceptorantagonist)inthetreatmentofneuropathicpainafter