rTMS in fibromyalgia · 2014. 3. 29. · patient. The participants were randomly assigned to 1 of the 2 following groups (random assignment 1:1): patients were assigned to receive

Laurent Boyer, MD,PhD*

Alix Dousset, MD*Philippe Roussel, MDNathalie Dossetto, PhDSerge Cammilleri, MD,

PhDVirginie Piano, MDStéphanie Khalfa, PhDOlivier Mundler, MD,

PhDAnne Donnet, MDEric Guedj, MD, PhD

Correspondence toProf. Guedj:[email protected]

Supplemental dataat Neurology.org

rTMS in fibromyalgiaA randomized trial evaluatingQoL and its brainmetabolic substrate

ABSTRACT

Objective: This double-blind, randomized, placebo-controlled study investigated the impact ofrepetitive transcranial magnetic stimulation (rTMS) on quality of life (QoL) of patients with fibro-myalgia, and its possible brain metabolic substrate.

Methods: Thirty-eight patients were randomly assigned to receive high-frequency rTMS (n 5 19) orsham stimulation (n5 19), applied to left primary motor cortex in 14 sessions over 10 weeks. Primaryclinical outcomes were QoL changes at the end of week 11, measured using the Fibromyalgia ImpactQuestionnaire (FIQ). Secondary clinical outcomes were mental and physical QoL component measuredusing the 36-Item Short Form Health Survey (SF-36), but also pain, mood, and anxiety. Resting-state[18F]-fluorodeoxyglucose-PET metabolism was assessed at baseline, week 2, and week 11. Whole-brain voxel-based analysis was performed to study between-group metabolic changes over time.

Results: At week 11, patients of the active rTMS group had greater QoL improvement in the FIQ (p 5

0.032) and in the mental component of the SF-36 (p 5 0.019) than the sham stimulation group. Nosignificant impact was found for other clinical outcomes. Compared with the sham stimulation group,patients of the active rTMS group presented an increase in right medial temporal metabolism betweenbaseline and week 11 (p , 0.001), which was correlated with FIQ and mental component SF-36concomitant changes (r 5 20.38, p 5 0.043; r 5 0.51, p 5 0.009, respectively). QoL improvementinvolved mainly affective, emotional, and social dimensions.

Conclusion: Our study shows that rTMS improvesQoL of patientswith fibromyalgia. This improvementis associated with a concomitant increase in right limbic metabolism, arguing for a neural substrate tothe impact of rTMS on emotional dimensions involved in QoL.

Classification of evidence: This study provides Class II evidence that rTMS compared with shamrTMS improves QoL in patients with fibromyalgia. Neurology® 2014;82:1–8

GLOSSARYBDI 5 Beck Depression Inventory; DSM-IV-R 5 Diagnostic and Statistical Manual of Mental Disorders, 4th edition, revised;FDG 5 [18F]-fluorodeoxyglucose; FIQ 5 Fibromyalgia Impact Questionnaire; MCS 5 Mental Composite Score; PCS 5 Phys-ical Composite Score; QoL 5 quality of life; rTMS 5 repetitive transcranial magnetic stimulation; SF-36 5 36-Item ShortForm Health Survey.

Fibromyalgia affects quality of life (QoL) to a large extent and more than other chronic pain con-ditions.1,2 Of interest, neuroimaging studies of fibromyalgia have suggested global dysfunction ofcentral pain processing.3–5 Perfusion abnormalities have been in particular described at rest (i.e.,without painful stimulation) with hyperperfusions of somatosensory cortex, supporting increasednociceptive perception, and with cortico-limbic hypoperfusions, supporting altered emotionalregulation.6 These hypofunctional brain regions have also been involved in mental processesrelated to self-awareness and awareness of others, and in a range of social cognitive abilities.7

Based on the hypothesis of central dysfunction,8 2 studies have assessed the effect of high-frequency repetitive transcranial magnetic stimulation (rTMS) on the left primary motor cortexand have reported a beneficial effect on pain and QoL.9,10 However, the mechanisms underlying the

*These authors contributed equally to this work.

From EA 3279–Self-perceived Health Assessment Research Unit (L.B.), School of Medicine, La Timone University, Marseille; Centre d’Evaluation et deTraitement de la Douleur (A. Dousset, P.R., N.D., V.P., A. Donnet) and Service Central de Biophysique et Médecine Nucléaire (S.C., O.M., E.G.),APHM, Hôpital de la Timone, Marseille; CERIMED (S.C., O.M., E.G.) and Institut de Neurosciences de la Timone, CNRS UMR 7289 (S.K., E.G.),Aix-Marseille Université, Marseille, France.

Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article.

© 2014 American Academy of Neurology 1

ª 2014 American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

mailto:[email protected]

http://neurology.org/


effectiveness of rolandic rTMS remain unclear.Because high-frequency stimulation is supposedto produce enhanced cortical responses (i.e.,facilitation),11,12 an activation of hypofunctionalsystems13,14 is expected, rather than an inhibitionof the somatosensory cortex. The clinical effectshould thus predominantly improve hypofunc-tional dimensions, affective, emotional, orsocial,15,16 which are involved in QoL, ratherthan directly affect pain nociception.

We designed a double-blind, randomized,and placebo-controlled trial to investigate theimpact of high-frequency rTMS over the leftprimary motor cortex on QoL in patients withfibromyalgia, and its brain metabolic substrate.

METHODS Study site and patient eligibility. This studywas conducted at La Timone University Hospital, in a specialized

pain treatment center, and in the Nuclear Medicine Department

(Marseille, France) from October 2008 to December 2010.

The inclusion criteria were as follows: age older than 18 years;

right-handed; diagnosis of fibromyalgia according to AmericanCollege

of Rheumatology criteria17; score of at least 4 on the average pain

intensity numerical scale of the Brief Pain Inventory18 at screening;

persistent pain for more than 6 months before enrollment; stable

treatment for more than 1 month before enrollment and throughout

the study; rTMS-naive; and native French speaking. At screening, all

patients underwent physical examination by the same pain specialist,

followed by laboratory/imaging tests if necessary. The exclusion criteria

were as follows: reduced capacity to consent; inflammatory rheumatic

disease, autoimmune disease, or other painful disorders that might

confound the assessment of fibromyalgia pain; current primary psy-

chiatric condition, including major depression or major personality

disorders according toDSM-IV-R criteria19—or a history of substance

abuse; neurologic disorders; and contraindications for rTMS and

[18F]-fluorodeoxyglucose (FDG)-PET, including history of seizures,

brain trauma, brain surgery, intracranial hypertension, a pacemaker

or othermetallic implants, and pregnancy/breastfeeding. Concomitant

medication for pain and sleep disorders was allowed, provided the dose

administered had been stable for at least 1 month before enrollment

and remained stable throughout the study.

Standard protocol approvals, registrations, and patientconsents. The patients were provided with both oral and written

information regarding the study before obtaining their informed

consent. The local ethics committee and the French drug and device

regulation agency approved this study. The international standard

randomized controlled trial number is NCT00697398.

Design. The present study was prospective, randomized, controlled,

double-blind, and monocentric. Figure e-1 on the Neurology® Web

site at Neurology.org displays the flowchart. Individuals were

randomized by a computer-generated list, which was maintained

centrally so no investigators knew the treatment allocation of any

patient. The participants were randomly assigned to 1 of the 2

following groups (random assignment 1:1): patients were assigned

to receive active rTMS or sham stimulation. QoL, depression,

anxiety, pain, and brain PET metabolism were assessed at 3 time

points: at randomization (baseline, T0), at 2 weeks (T1), and at 11

weeks (1 week after last stimulation, T2).

rTMS protocol. The stimulation protocol consisted of 14 ses-

sions over 10 weeks: an “induction phase” of 10 sessions over 2

weeks followed by a “maintenance phase” of 4 sessions (1 session

at weeks 4, 6, 8, and 10).9,10 Sham stimulation was conducted

with a sham coil of identical size, color, and shape, emitting a

sound similar to that emitted by the active coil. Stimulations were

administered by the same technologist. Patients and clinical raters

were blinded to treatment. The characteristics of the magnetic

stimulation are presented in appendix e-1.

Evaluation criteria. Primary criterion. The primary evalua-

tion criterion was QoL change from baseline to T2. QoL was spe-

cifically assessed using the French version of the Fibromyalgia

Impact Questionnaire (FIQ).20 The total score of the FIQ (range

0–100) provides an estimation of the impact of fibromyalgia,

with higher scores indicating lower QoL levels.

Secondary criteria. QoL was also assessed using a generic scale,

the Medical Outcomes Study 36-Item Short Form Health Survey

(SF-36),21 to identify the 2 QoL components, namely, the Physical

Composite Score (PCS) and Mental Composite Score (MCS). Scores

range from 0 to 100, with higher scores indicating higher QoL levels.

Pain was measured using a self-reported average pain intensity

scale over the last 24 hours (numerical scale from 0 5 no pain to

105 maximal pain imaginable), the number of tender points (of 18

points in total), and pressure pain threshold (i.e., minimum force

applied that induces pain).

Depression was assessed using the self-administered Beck Depres-

sion Inventory (BDI).22 The BDI score range is 0 to 63, with higher

scores indicating greater depression. Anxiety was assessed using the

Hospital Anxiety and Depression Scale23; scores range from 0 to 21,

with higher scores indicating more severe anxiety symptoms.

Brain metabolism was studied with FDG-PET/CT imaging.

Brain data analysis is detailed in appendix e-2. Statistical maps

were thresholded at p 5 0.001 and corrected for extent to 16

voxels (4 full width at half maximum of the Gaussian filter).

Statistical analyses. In the 2 previous trials that have assessed

the impact of rTMS in fibromyalgia,9,10 differences in the FIQ score

were found between active rTMS and sham stimulation groups.

Guided by these results, the sample size was initially determined to

obtain an 80% power to detect a 10-point difference in QoL at T2 as

evaluated by the FIQ. With a significant p value of 0.05, these

calculations showed that a total of 30 patients were needed;

considering a potential 20% of patients being lost to follow-up,10

a total of 38 patients would need to be included.

Clinical analyses were performed on the intent-to-treat popula-

tion. Changes in outcomes between T0 and T1/T2 were compared

across the 2 groups using Mann-Whitney U tests for continuous or

ordinal variables, and x2 or Fisher exact tests for frequencies. Changes

in outcome were also analyzed with mixed-effect models for contin-

uous variables or nonparametric analysis of longitudinal data for ordi-

nal variables (number of tender points and average daily pain).24,25

The models included the following: QoL scores over time as the

outcome variable (FIQ, SF-36 PCS and MCS); BDI scores as cova-

riates to isolate the effect of rTMS on QoL, independently of depres-

sion; time as a categorical variable, study group, time3 study group

and BDI score3 study group as fixed effects; and patient as random

effect. In addition to intention-to-treat analysis (sensitivity analysis),

complementary per-protocol analyses were undertaken on significant

clinical data.

Finally, to prevent unnecessary radiation exposure, brain imaging

at T1/T2 was performed only in patients who completed the rTMS

treatment. Neuroimaging analyses were based on Statistical Paramet-

ric Mapping. We then looked for associations in metabolic and clin-

ical changes over time using Spearman correlation tests.

Statistical significance was set at p , 0.05. Statistical analyses

were performed using SPSS for Windows (version 17.0; SPSS

2 Neurology 82 April 8, 2014


http://www.neurology.org/

Inc., Chicago, IL) and R Software Package for the Nonparametric

Analysis of Longitudinal Data in Factorial Experiments.24,25

Primary research question and classification of evidence.We hypothesized that high-frequency rTMS would have a

predominant effect on QoL rather than on pain. This study

provides Class II evidence that rTMS compared with sham

rTMS improves QoL in patients with fibromyalgia.

RESULTS Participants. Table 1 shows baseline data forthe 38 patients randomly assigned to the active rTMSgroup or the sham stimulation group. The 2 groups werewell balanced for pretreatment characteristics, includingbrain PET metabolism. All patients completed theinduction phase, but 9 (23.7%) were excluded duringthe maintenance phase (3 in the active rTMS group and

6 in the sham rTMS group) (figure e-1). The reasonsincluded 5 intercurrent medical conditions forcing atherapeutic shift (1 cystitis/pyelonephritis and 1 lumbagoin the active rTMS group; 1 tracheitis, 1 kidney stonecolic, and 1 headache in the sham rTMS group); 2spontaneous changes in pain automedication (1 in eachgroup); and lack of efficacy in 2 patients who decided tostop the protocol (sham rTMS group). No seizures orother side effects occurred during the follow-up study.

Effects of rTMS on FIQ and secondary clinical outcomes.

At T2, the active rTMS group had a greater QoLimprovement than did the control group for the pri-mary outcome (FIQ score: 29.6 6 16.7 vs 12.0 6

9.3 points, p 5 0.032) and for one secondary

Table 1 Baseline characteristics of the study participantsa

Variable Active rTMS group (n 5 19) Sham stimulation group (n 5 19) p Value

Sex, female, n (%) 19 (100) 18 (94.7) 0.311

Age, y 49.1 6 10.6 47.7 6 10.4 0.759

Single, n (%) 3 (15.8) 2 (10.5) 0.631

Secondary school level, n (%) 9 (47.4) 13 (68.4) 0.189

Employment, n (%) 12 (63.2) 13 (68.4) 0.732

Duration of fibromyalgia-related pain, y 3.7 6 4.5 3.6 6 3.8 0.723

Medications taken before intervention, n (%)

Antidepressants 9 (47.4) 12 (63.2) 0.328

Anticonvulsants 7 (36.8) 5 (26.3) 0.485

Benzodiazepines 6 (31.6) 10 (52.6) 0.189

NSAIDs 1 (5.3) 1 (5.3) 1.000

Muscle relaxants 2 (10.5) 2 (10.5) 1.000

Opioid analgesics 2 (10.5) 1 (5.3) 0.547

Nonopioid analgesics 11 (57.9) 14 (73.7) 0.305

FIQ scoreb 60.0 6 11.6 64.1 6 11.6 0.343

SF-36 scorec

Physical Composite Score 29.9 6 7.5 32.4 6 5.9 0.327

Mental Composite Score 39.6 6 11.4 34.0 6 9.3 0.137

BDId 9.1 6 5.9 11.7 6 8.1 0.282

HADS anxiety scoree 10.8 6 2.4 9.4 6 2.4 0.103

No. of tender points, 0–18 17.4 6 1.0 17.9 6 0.3 0.092

Average daily pain, 0–10 6.5 6 1.6 6.5 6 2.0 0.552

Pressure pain threshold, kPa

Right arm 83.7 6 29.5 84.2 6 35.3 0.965

Left arm 78.9 6 29.0 73.2 6 27.1 0.626

Medial temporal lobe metabolismf 33.5 6 3.8 34.2 6 2.7 0.511

Abbreviations: BDI 5 Beck Depression Inventory; FIQ 5 Fibromyalgia Impact Questionnaire; HADS 5 Hospital Anxiety andDepression Scale; NSAID 5 nonsteroidal anti-inflammatory drug; rTMS 5 repetitive transcranial magnetic stimulation;SF-36 5 36-Item Short Form Health Survey.aValues are means 6 SD unless otherwise noted.bScores range from 0 to 100, with higher scores indicating more severe symptoms.c The Medical Outcomes Study SF-36. Scores range from 0 to 100, with higher scores indicating better health status.dScores range from 0 to 63, with higher scores indicating more severe depressive symptoms.eAnxiety scores range from 0 to 21, with higher scores indicating more severe anxiety symptoms.f Per-protocol analysis: n 5 16 in the active rTMS group and n 5 13 in the sham stimulation group.

Neurology 82 April 8, 2014 3


outcome (SF-36 MCS: 15.0 6 6.9 vs 21.6 6 7.6,p 5 0.019) (table 2, figure 1). A difference in FIQscore and SF-36 MCS between groups from baselineto T2 was confirmed using mixed-effect models (p 50.033 and p , 0.001, respectively) and also usingper-protocol analysis (p 5 0.032 and p , 0.001,respectively). The QoL was more improved in therTMS group for all dimensions of the SF-36, butonly the mental health dimension was significant(p 5 0.045). The bodily pain dimension was lessaffected by rTMS than other dimensions involvingemotional or social issues (appendix e-3). Nodifferences for SF-36 PCS, and for other clinicaloutcomes (pain, mood, and anxiety), were observedover time from baseline between the 2 groups(p values .0.05). At T0, BDI scores indicated milddepression in both groups (.9 points), and at T2 thescores remained stable in the control group whereas theydecreased by 2 points in the active rTMS group,indicating minimal depression. However, the differencewas statistically nonsignificant (p5 0.346). In the sameway, pressure pain threshold of the right arm at T2 wasmore increased in the rTMS group (110.5) than in thecontrol group (10.5), but again this difference was notsignificant (p 5 0.528).

Effect of rTMS on brain PET metabolism. In comparisonto the sham stimulation group, the active rTMS grouppresented an increase in right medial temporal metabo-lism (hippocampus, parahippocampal and fusiformgyrus, Brodmann area 20; k 5 89; t score 5 3.85;p, 0.001; figure 2) between T0 and T2. It is interest-ing that this metabolic increase relative to baseline wasalready present at T1 (p 5 0.002), before the improve-ment of QoL at T2.

We then looked for Spearman correlations betweenthis medial temporal metabolism and QoL scores.Right medial temporal metabolism and QoL variedin the same direction between T2 and T0 (FIQ: r 520.38, p5 0.043; SF-36MCS: r5 0.51, p5 0.009;figure 3), especially for nonphysical dimensions, suchas the feel good (r 5 20.41, p 5 0.031), work missed(r 5 20.58, p 5 0.039), depression (r 5 20.37,p 5 0.046), fatigue (r 5 20.50, p 5 0.006), rested(r520.40, p5 0.034), and stiffness (r520.45, p50.014) dimensions of the FIQ and the role-emotionaldimension of the SF-36 (r 5 0.40, p 5 0.040).

DISCUSSION This randomized, double-blind, sham-controlled study showed that high-frequency rTMSover the left primary motor cortex had a delayedpositive impact on patients’ QoL after 11 weeks oftreatment, without effect on pain. Above all, thisneuroimaging study showed that the therapeuticeffect of rTMS over the left motor cortex wascorrelated with an increase in right medial temporalmetabolism. This whole-brain voxel-based report

Table 2 Changes in primary and secondary outcomesa

Variable

Mean change from baseline

p ValueActive rTMSgroup

Sham stimulationgroup

FIQ scorea

Week 2 0.3 6 18.2 1.3 6 9.5 0.671

Week 11 29.6 6 16.7 2.0 6 9.3 0.032b

SF-36 scorec

Physical Composite Score

Week 2 0.3 6 8.8 0.9 6 5.3 0.666

Week 11 1.4 6 9.0 0.4 6 4.8 0.874

Mental Composite Score

Week 2 2.1 6 9.4 20.4 6 9.4 0.264

Week 11 5.0 6 6.9 21.6 6 7.6 0.019b

BDId

Week 2 21.3 6 2.7 20.5 6 4.0 0.320

Week 11 21.9 6 2.8 20.1 6 4.4 0.346

HADS anxiety scoree

Week 2 0.3 6 2.6 20.8 6 1.9 0.244

Week 11 0.4 6 1.7 0.5 6 2.3 0.724

Tender points, 0–18

Week 2 20.2 6 2.2 21.4 6 4.4 0.215

Week 11 21.5 6 2.7 23.7 6 6.5 0.613

Average daily pain, 0–10

Week 2 0.1 6 1.8 20.2 6 2.7 0.485

Week 11 20.3 6 1.6 21.5 6 3.1 0.369

Pressure pain threshold, kPa

Right arm

Week 2 3.2 6 25.0 4.7 6 33.1 0.870

Week 11 10.5 6 28.8 0.5 6 45.4 0.528

Left arm

Week 2 21.6 6 27.1 22.1 6 37.5 0.814

Week 11 21.1 6 28.8 27.9 6 52.9 0.509

Medial temporal lobe metabolismf

Week 2 6.2 6 3.8 1.0 6 3.6 0.002b

Week 11 10.3 6 5.9 22.4 6 5.3 ,0.0001b

Abbreviations: BDI 5 Beck Depression Inventory; FIQ 5 Fibromyalgia Impact Questionnaire;HADS 5 Hospital Anxiety and Depression Scale; rTMS 5 repetitive transcranial magneticstimulation; SF-36 5 36-Item Short Form Health Survey.Values are means 6 SD.aScores range from 0 to 100, with higher scores indicating more severe symptoms.bStatistically significant values.cMedical Outcomes Study SF-36. Scores range from 0 to 100, with higher scoresindicating better health status.dScores range from 0 to 63, with higher scores indicating more severe depressivesymptoms.eAnxiety scores range from 0 to 21, with higher scores indicating more severe anxietysymptoms.f Per-protocol analysis: n 5 16 in the active rTMS group and n 5 13 in the sham stimulationgroup.



corroborates the central hypothesis in fibromyalgia,8

while previous studies (based on region of interest)failed to demonstrate brain PET metabolism changesin these patients.26

QoL was improved after rTMS maintenance onboth the FIQ and mental component of the SF-36.The focus on these 2 distinct questionnaires permitsa precise analytical description, with predominantQoL improvement involving affective (depressionand fatigue), emotional (role-emotional), and social(work missed) dimensions. In line with this finding,the BDI scores decreased in the active rTMS group.The absence of statistical significance may be causedby a lack of sensitivity to change of this scale, as pre-viously reported.27 However, the effect on QoL wasnot immediate; it appeared only at the end of themaintenance phase. Despite this positive outcome,there was no effect on pain. The rTMS effects maythus reflect an improvement involving the emotionaldimension associated with pain, rather than a direct

Figure 1 Mean changes in quality of life and resting-state brain metabolism at 2 and 11 weeks

(A) FIQ total score (p 5 0.032). (B) Total SF-36 MCS (p 5 0.019). (C) Total SF-36 PCS (p 5 0.874). (D) Medial temporalmetabolism (p , 0.001). FIQ 5 Fibromyalgia Impact Questionnaire; MCS 5 Mental Composite Score; PCS 5 PhysicalComposite Score; rTMS 5 repetitive transcranial magnetic stimulation; SF-36 5 36-Item Short Form Health Survey.

Figure 2 Anatomical localization of FDG-PET findings (p < 0.001)

Increase in medial temporal metabolism (hippocampus, parahippocampal and fusiform gyrus) inthe active repetitive transcranial magnetic stimulation group. FDG 5 [18F]-fluorodeoxyglucose.



affect on the sensory component of pain.10 Our find-ings, however, differ substantially from those of the 2previous studies, which showed immediate effects onboth QoL and pain.9,10 Considering the small samplesize of these studies, one cannot exclude that the pop-ulations have different characteristics and thus differentresponse profiles. Moreover, the stimulation protocolswere not exactly the same.

Besides these methodologic considerations, the fail-ure to detect between-group differences in pain mayhave several explanations. One hypothesis is that rTMShas an influence on the psychological dimensionsinvolved in QoL, without effect on neural processingof pain. In our study, improvement of QoL would berelated only to a better perception of health, but notto a pain decrease. Of interest, a recent study has sug-gested that 2 segregated mechanisms were involved inthe neural processing of pain and of negative affect.28

However, we may also hypothesize that our stimulationprotocol was too short to detect a global pain improve-ment. rTMS-induced analgesia, rather than actingdirectly on pain, may be mediated by the translationof changes in emotional processing associated withglobal pain, thus requiring a delay in action. In supportof this hypothesis, a recent study suggested that cata-strophizing might precede changes in pain response.29

Our results are, however, not surprising from a theo-retical perspective. In accordance with the metabolismincrease found, high-frequency rTMS has been associatedwith facilitatory effects on cortical excitability.12,30 Con-versely, the metabolic effects of rTMS involved contra-lateral interconnected brain structures, as reported inprevious studies,31,32 but not the primary targeted corticalarea. This absence of local effect of rTMS appears sur-prising, but has also been previously reported.19 Facilita-tory effects may become weaker when the targetedcortical area is already activated, as expected for the soma-tosensory cortex in patients with fibromyalgia.5 It is inter-esting that the role of the limbic system, and that of theright medial temporal cortex, have been described inemotional modulation, in particular in modulation ofthe emotional aspects associated with pain.33 However,the temporal lobe has also been associated with socialcognition, especially in theory of mind abilities.7 Simi-larly, a recent study reported that the superior temporalsulcus was involved in the functional substrate underlyingsocial functioning of QoL.34 Moreover, neural connec-tions have been reported between this temporal area andthe limbic system.35 We may thus hypothesize that theeffects of high-frequency rTMS over the hemisphere-dominant motor cortex mainly activate interconnectedemotional and social systems, resulting in improvedQoL.

The increase in medial temporal lobe metabolismbefore the improvement of QoL could have clinical im-plications. Biomarkers such as neuroimaging could openinteresting perspectives,36 to guide individual therapeu-tic strategy, especially to early identified responders andnonresponders to rTMS.5 Early detection of nonres-ponders to rTMSmay then allow proposing more adap-ted treatment. Indeed, despite the safety and tolerance ofrTMS found in our study, a main concern of treatmentburden remains, especially in patients with chronic painconditions.

The potential mechanism underlying high-frequencyrTMS over the primary motor cortex leads us to discussthe relevance of using further rTMS protocols, in combi-nation or alone, to optimize clinical effects. Because high-frequency stimulation has mainly produced an activationof right limbic structures, it would be preferable todirectly stimulate an ipsilateral reachable cortex, highlyinterconnected with the deep limbic cortex, as proposedin depression for the dorsolateral prefrontal cortex37 andwith encouraging results on affective dimensions. How-ever, because of inhibitory effects of low-frequency stim-ulation on cortical excitability,38 low-frequency rTMSover the sensorimotor cortex should decrease nociceptiveperception and then have direct analgesic effects in fibro-myalgia, as reported in dystonia for the low-frequencyrTMS of premotor cortex on painful muscular spasms.39

Finally, rTMS remains a constraining treatment,probably not appropriate for all patients because ofspecific contraindications as well as the heterogeneity

Figure 3 Correlations between right medial temporal lobe metabolism and FIQ(A)/SF-36 MCS (B) changes between baseline and week 11

FIQ 5 Fibromyalgia Impact Questionnaire; MCS 5 Mental Composite Score; SF-36 5 36-Item Short Form Health Survey.



of rTMS response. The absence of placebo effect canbe surprising. We hypothesize that the length and theburden of our protocol exhausted the patients, whichcould explain the absence of placebo effect. Thelength of our protocol, the allocated resources in per-sonnel and equipment, the patients’ compliance, andthe absence of pain improvement should be consid-ered in future studies to optimize the clinical practiceof rTMS in patients with fibromyalgia.

This study is limited by its small sample size(n 5 38), although it is close to the only other studyon the efficacy of high-frequency rTMS maintenanceover the motor cortex in fibromyalgia (n 5 40).10 Inaddition, 9 patients did not complete the mainte-nance phase, reducing our sample to 29 patients. Thisproportion of those withdrawn (23.7%) was similarto that of the previous study (25.0%).10 Moderateefficacy may have been missed, especially for pain.This issue is of importance in fibromyalgia, for whichtreatments are known to have moderate effects.40

Replication with a larger sample size is needed.Although patients were blinded to the treatment con-

dition, the rTMS administrator who placed the active orsham stimulator in position could not be blinded to thetreatment. To overcome this potential bias, she was notinvolved in the recruitment and evaluation of patients. Ablinded rater was used for pain measures and the occur-rence of adverse effects. Moreover, all patients wererTMS-naive, preventing them from recognizing thesham or active coils. Recent studies recommended thatinvestigators test for the success of blinding, which wasnot done in the present trial.

Overall, our study shows that active rTMS im-proves QoL of patients with fibromyalgia. Thisimprovement is associated with a concomitantincrease in right limbic metabolism, arguing for aneural substrate to the impact of rTMS on the emo-tional dimension involved in QoL.

AUTHOR CONTRIBUTIONSL.B.: drafting/revising the manuscript for content, including medical writing for

content, analysis or interpretation of data, statistical analysis. A. Dousset: draft-

ing/revising the manuscript for content, including medical writing for content,

analysis or interpretation of data. P.R.: study concept or design, analysis or

interpretation of data, acquisition of data, study supervision or coordination,

obtaining funding. N.D.: study concept or design, acquisition of data. S.C.:

study concept or design, acquisition of data, obtaining funding. V.P.: analysis

or interpretation of data. S.K.: drafting/revising the manuscript for content,

including medical writing for content, analysis or interpretation of data.

O.M.: study concept or design, obtaining funding. A. Donnet: drafting/revising

the manuscript for content, including medical writing for content, analysis or

interpretation of data, study supervision or coordination. E.G.: drafting/revising

the manuscript for content, including medical writing for content, study con-

cept or design, analysis or interpretation of data, acquisition of data, statistical

analysis, study supervision or coordination, obtaining funding.

STUDY FUNDINGSupported by Inserm (Centre d’Investigation Clinique, CIC, Hôpital de

la Conception, Marseille) and AP-HM (AORC 2008/01). International

Standard Randomized Controlled Trial Number: NCT00697398.

DISCLOSUREThe authors report no disclosures relevant to the manuscript. Go to

Neurology.org for full disclosures.

Received March 1, 2013. Accepted in final form January 6, 2014.

REFERENCES1. Hoffman DL, Dukes EM. The health status burden of

people with fibromyalgia: a review of studies that assessed

health status with the SF-36 or the SF-12. Int J Clin Pract

2008;62:115–126.

2. Mas AJ, Carmona L, Valverde M, Ribas B. Prevalence and

impact of fibromyalgia on function and quality of life in in-

dividuals from the general population: results from a nation-

wide study in Spain. Clin Exp Rheumatol 2008;26:519–526.

3. Gracely RH, Petzke F, Wolf JM, Clauw DJ. Functional mag-

netic resonance imaging evidence of augmented pain process-

ing in fibromyalgia. Arthritis Rheum 2002;46:1333–1343.

4. Guedj E. Neuroimaging findings in fibromyalgia: what

clinical impact? Joint Bone Spine 2009;76:224–226.

5. Guedj E, Cammilleri S, Niboyet J, et al. Clinical correlate

of brain SPECT perfusion abnormalities in fibromyalgia.

J Nucl Med 2008;49:1798–1803.

6. Guedj E, Taieb D, Cammilleri S, et al. 99mTc-ECD brain

perfusion SPECT in hyperalgesic fibromyalgia. Eur J Nucl

Med Mol Imaging 2007;34:130–134.

7. Giovagnoli AR, Franceschetti S, Reati F, et al. Theory

of mind in frontal and temporal lobe epilepsy: cognitive

and neural aspects. Epilepsia 2011;52:1995–2002.

8. Desmeules JA, Cedraschi C, Rapiti E, et al. Neurophysi-

ologic evidence for a central sensitization in patients with

fibromyalgia. Arthritis Rheum 2003;48:1420–1429.

9. Passard A, Attal N, Benadhira R, et al. Effects of unilateral

repetitive transcranial magnetic stimulation of the motor

cortex on chronic widespread pain in fibromyalgia. Brain

2007;130:2661–2670.

10. Mhalla A, Baudic S, Ciampi de Andrade D, et al. Long-term

maintenance of the analgesic effects of transcranial magnetic

stimulation in fibromyalgia. Pain 2011;152:1478–1485.

11. Berardelli A, Inghilleri M, Rothwell JC, et al. Facilitation

of muscle evoked responses after repetitive cortical stimu-

lation in man. Exp Brain Res 1998;122:79–84.

12. Pascual-Leone A, Valls-Sole J, Wassermann EM, Hallett M.

Responses to rapid-rate transcranial magnetic stimulation of

the human motor cortex. Brain 1994;117(pt 4):847–858.

13. Garcia-Larrea L, Peyron R, Mertens P, et al. Electrical stimu-

lation of motor cortex for pain control: a combined PET-scan

and electrophysiological study. Pain 1999;83:259–273.

14. Lefaucheur JP. Use of repetitive transcranial magnetic stimu-

lation in pain relief. Expert Rev Neurother 2008;8:799–808.

15. Gracely RH, Geisser ME, Giesecke T, et al. Pain cata-

strophizing and neural responses to pain among persons

with fibromyalgia. Brain 2004;127:835–843.

16. Montoya P, Larbig W, Braun C, Preissl H, Birbaumer N.

Influence of social support and emotional context on pain

processing and magnetic brain responses in fibromyalgia.

Arthritis Rheum 2004;50:4035–4044.

17. Wolfe F, Clauw DJ, Fitzcharles MA, et al. The American

College of Rheumatology preliminary diagnostic criteria

for fibromyalgia and measurement of symptom severity.

Arthritis Care Res 2010;62:600–610.

18. Cleeland CS, Ryan KM. Pain assessment: global use of the

Brief Pain Inventory. Ann Acad Med Singapore 1994;23:

129–138.




19. American Psychiatric Association. Diagnostic and Statistical

Manual of Mental Disorders: DSM-IV-R. Washington,

DC: American Psychiatric Association; 2000.

20. Perrot S, Dumont D, Guillemin F, Pouchot J, Coste J.

Quality of life in women with fibromyalgia syndrome: val-

idation of the QIF, the French version of the fibromyalgia

impact questionnaire. J Rheumatol 2003;30:1054–1059.

21. Ware JE Jr, Sherbourne CD. The MOS 36-Item Short-

Form Health Survey (SF-36): I: conceptual framework and

item selection. Med Care 1992;30:473–483.

22. Beck AT, Epstein N, Brown G, Steer RA. An inventory

for measuring clinical anxiety: psychometric properties.

J Consult Clin Psychol 1988;56:893–897.

23. Zigmond AS, Snaith RP. The hospital anxiety and depres-

sion scale. Acta Psychiatr Scand 1983;67:361–370.

24. Brunner E, Domhof S, Langer F. Nonparametric Analysis

of Longitudinal Data in Factorial Experiments. New York:

Wiley; 2002.

25. Noguchi K, Gel YR, Brunner E, Konietschke F. nparLD:

an r software package for the nonparametric analysis of

longitudinal data in factorial experiments. J Stat Softw

2012;50:1–23.

26. Yunus MB, Young CS, Saeed SA, Mountz JM, Aldag JC.

Positron emission tomography in patients with fibromyal-

gia syndrome and healthy controls. Arthritis Rheum 2004;

51:513–518.

27. Schneibel R, Brakemeier EL, Wilbertz G, Dykierek P,

Zobel I, Schramm E. Sensitivity to detect change and the

correlation of clinical factors with the Hamilton Depression

Rating Scale and the Beck Depression Inventory in

depressed inpatients. Psychiatry Res 2012;198:62–67.

28. Jensen KB, Petzke F, Carville S, et al. Anxiety and depres-

sive symptoms in fibromyalgia are related to poor percep-

tion of health but not to pain sensitivity or cerebral

processing of pain. Arthritis Rheum 2010;62:3488–3495.

29. Campbell CM, McCauley L, Bounds SC, et al. Changes in

pain catastrophizing predict later changes in fibromyalgia

clinical and experimental pain report: cross-lagged panel

analyses of dispositional and situational catastrophizing.

Arthritis Res Ther 2012;14:R231.

30. Hallett M. Transcranial magnetic stimulation: a primer.

Neuron 2007;55:187–199.

31. Bestmann S, Baudewig J, Siebner HR, Rothwell JC,

Frahm J. Functional MRI of the immediate impact of

transcranial magnetic stimulation on cortical and subcor-

tical motor circuits. Eur J Neurosci 2004;19:1950–1962.

32. Yoo WK, You SH, Ko MH, et al. High frequency rTMS

modulation of the sensorimotor networks: behavioral changes

and fMRI correlates. Neuroimage 2008;39:1886–1895.

33. Peyron R, Garcia-Larrea L, Gregoire MC, et al. Haemody-

namic brain responses to acute pain in humans: sensory and

attentional networks. Brain 1999;122(pt 9):1765–1780.

34. Boyer L, Richieri R, Faget C, et al. Functional involvement of

superior temporal sulcus in quality of life of patients with

schizophrenia. Psychiatry Res 2012;202:155–160.

35. Kleinhans NM, Richards T, Sterling L, et al. Abnormal

functional connectivity in autism spectrum disorders dur-

ing face processing. Brain 2008;131:1000–1012.

36. Guedj E, Cammilleri S, Colavolpe C, et al. Predictive

value of brain perfusion SPECT for ketamine response

in hyperalgesic fibromyalgia. Eur J Nucl Med Mol Imag-

ing 2007;34:1274–1279.

37. Richieri R, Boyer L, Farisse J, et al. Predictive value of

brain perfusion SPECT for rTMS response in pharmacor-

esistant depression. Eur J Nucl Med Mol Imaging 2011;

38:1715–1722.

38. Modugno N, Nakamura Y, MacKinnon CD, et al. Motor

cortex excitability following short trains of repetitive mag-

netic stimuli. Exp Brain Res 2001;140:453–459.

39. Lefaucheur JP, Fenelon G, Menard-Lefaucheur I,

Wendling S, Nguyen JP. Low-frequency repetitive TMS

of premotor cortex can reduce painful axial spasms in gen-

eralized secondary dystonia: a pilot study of three patients.

Neurophysiol Clin 2004;34:141–145.

40. Arnold LM. Strategies for managing fibromyalgia. Am J

Med 2009;122:S31–S43.

Complete the AAN 2014 Neurology Compensation andProductivity Survey by May 9

The AAN launched its second annual Neurology Compensation and Productivity Survey in Marchand needs practicing US members and their practices to contribute their data. It is critical that allUS neurologists and practice managers participate in the survey to ensure the most accurate andauthoritative data representing the US neurology landscape. Visit AAN.com/view/2014NeuroSurveyto review preparation documents, including an FAQ and Quick Start Guide. Complete the surveyby May 9 and get free access to the online results and the Neurology Compensation and ProductivityReport, available in early July 2014. The cost to access the data and report for nonparticipants is$600 for AAN members and $1200 for nonmembers.



Documents

rTMS in fibromyalgia · 2014. 3. 29. · patient. The participants were randomly assigned to 1 of the 2 following groups (random assignment 1:1): patients were assigned to receive