Complementary Therapies in Medicine (2014) 22, 731—742

Available online at www.sciencedirect.com

jo ur nal home p ag e: www.elsev ierhea l th .com/ journa ls /c t im

Homeopathy in the treatment offibromyalgia—–A comprehensiveliterature-review and meta-analysis

Katja Boehma, Christa Raaka,∗, Holger Cramerb, Romy Laucheb,Thomas Ostermanna,∗

a Institute for Integrative Medicine, Witten/Herdecke University, Germanyb Department of Internal and Integrative Medicine, Kliniken-Essen-Mitte, Faculty of Medicine, University ofDuisburg-Essen, GermanyAvailable online 28 June 2014

KEYWORDSFibromyalgia;Homeopathy;Review;Meta-analysis

SummaryBackground: Coping with the complex nature of fibromyalgia symptoms (FMS) still remains achallenge for patients. Taking into account the possible adverse events of pharmacologicaltreatments patients often seek additional treatments for the management of fibromyalgia andturn towards complementary and alternative medicine (CAM).Objective: In this review, we aimed to investigate the current state of literature of homeopathyin the treatment of FMS.Methods: We searched Medline, the Cochrane Register of Controlled Trials, Embase, AMED,PsycInfo and CAMbase for the terms ‘‘fibromyalgia AND homeopath$’’ through February 2013.In addition we searched Google Scholar, the library of the Carstens Foundation and that ofthe Deutsche Homöopathische Union (DHU). Standardized mean differences (SMD) with 95%confidence intervals (CI) were calculated and meta-analyzed using the generic inverse variancemethod.Results: We found 10 case-reports, 3 observational studies, 1 non-randomized and 4 ran-domized controlled trials (RCTs) on homeopathy for fibromyalgia. Both case reports andobservational studies are naturally predominated by the use of qualitative and not validatedoutcome measures. Meta-analyses of CCTs revealed effects of homeopathy on tender pointcount (SMD = −0.42; 95%CI −0.78, −0.05; P = 0.03), pain intensity (SMD = −0.54; 95%CI −0.97,−0.10; P = 0.02), and fatigue (SMD = −0.47; 95%CI −0.90, −0.05; P = 0.03) compared to placebo.

Conclusion: The results of the studies as well as the case reports define a sufficient basis for dis-cussing the possible benefits of homeopathy for patients suffering from fibromyalgia syndrome although any conclusions based on the results of this review have to be regarded as preliminary. © 2014 Elsevier Ltd. All rights re

∗ Corresponding authors. Tel.: +49 2330623643.E-mail addresses: christa.raak@uni-wh.de (C. Raak), thomas.osterma

http://dx.doi.org/10.1016/j.ctim.2014.06.0050965-2299/© 2014 Elsevier Ltd. All rights reserved.

served.

nn@uni-wh.de (T. Ostermann).

7

C

I

TcavFrsfis

eEetp1Wrisiof

btaaatf

32 K. Boehm et al.

ontents

Introduction.............................................................................................................. 732Methods.................................................................................................................. 733

Search strategy...................................................................................................... 733Study selection...................................................................................................... 733

Inclusion/exclusion criteria................................................................................... 733Selection process ............................................................................................. 733

Statistical analysis................................................................................................... 733Quality assessment .................................................................................................. 733

Results ................................................................................................................... 733Characteristics of included studies .................................................................................. 734

Case reports .................................................................................................. 734Uncontrolled clinical trials.................................................................................... 734Controlled clinical trial ....................................................................................... 735Randomized controlled trials ................................................................................. 735

Meta-analysis........................................................................................................ 736Quality assessment .................................................................................................. 736

Discussion................................................................................................................ 736Meta-analysis........................................................................................................ 740Single cases ......................................................................................................... 740Strengths and limitations............................................................................................ 740Implications for further research .................................................................................... 741

Conclusions .............................................................................................................. 741Conflict of interest statement............................................................................................ 741

Funding ................................................................................................................ 741References ............................................................................................................. 741

ntroduction

he fibromyalgia syndrome (FMS) is a condition defined byhronic widespread pain, fatigue, cognitive disturbancesnd sleep disorders.1,2 Patients with FMS also experiencearious somatic symptoms and psychological distress.1,2

ibromyalgia is a frequent comorbidity alongside otherheumatologic conditions. Thus patients experience sub-tantial disabilities and often report a negative impact ofbromyalgia on their quality of life, mood, anxiety, depres-ion and self-esteem.3

According to a recent epidemiological study of Brancot al.4 fibromyalgia affects about 1.4—3.7% of adults inurope depending on the country. Epidemiological studiesstimated how many people in the general population meethe FM-criteria at the time and found that with a Euro-ean point prevalence of 2.9% this leads to a total of about5 million people in Europe suffering from fibromyalgia.4

ith respect to the costs a recent study of Berger et al.5

eports three times higher healthcare costs over 12 monthsn fibromyalgia patients compared to a matched patientample in the US. These results are comparable with find-ng of the same research group in German GPs: compared tother primary care patients, fibromyalgia patients countedor twice as many GP visits, referrals and sick notes.6

In conventional medical practice, fibromyalgia is treatedy using a wide range of symptom specific pharmacologicalherapies, including antidepressants, opioids, non-steroidalnti-inflammatory drugs, sedatives, muscle relaxants, and

by a team of UK researchers in 2008 included antide-pressants, analgesics, and ‘‘other pharmacological’’ andexercise, cognitive behavioural therapy, education, dietaryinterventions and ‘‘other non-pharmacological’’.32 Treat-ment by opioids (except Tramadol) was not recommendedby recent evidence-based guidelines.

However, coping with the complex nature of fibromyal-gia symptoms still remains a challenge for patients. Takinginto account the possible adverse events of pharmacolog-ical treatments patients often seek additional treatmentsfor the management of fibromyalgia and turn towards com-plementary and alternative medicine (CAM). According to asurvey by Wahner-Roedler et al.7 89% of patients referredto a fibromyalgia treatment programme at a tertiary carecentre had used at least some type of CAM therapy dur-ing the previous 6 months including exercise therapy (48%),spiritual healing and prayers (45%), massage therapy (44%),chiropractic treatments (37%), or vitamins and minerals(35—25%). Nevertheless the evidence base for many of thosetherapeutic options for fibromyalgia is quite sparse and themethodological quality of clinical studies often is low.

A recent overview of Terry et al.,8 a meta-analysis ofLanghorst et al.,9 and a qualitative review of Baranowskyet al.10 not only found positive results for hydrotherapy andspa therapy but also for homeopathy. According to Perryet al.8 existing RCTs in homeopathy suggested results infavour of homeopathy which up to now have not been sum-marized by means of a meta-analysis. Moreover, results fromobservational studies or even case reports have not been

nti-epileptics.33 Non-pharmaceutical treatments includeerobic exercises, physical therapies, massage, and cogni-ive behavioural therapy. Evidence-based recommendationsor the management of fibromyalgia syndrome identified

c

hh

ollected to complement these findings.

Thus, in the following review we aimed to compre-

ensively investigate the current state of literature foromeopathic interventions in the treatment of patients

ps

S

Mcrtesttm9gaaefs0

n

dpetatm

po

Q

QaTbcvdbsrP

R

Iv

Homeopathy in the treatment of fibromyalgia

suffering from fibromyalgia and examine all reports fromsingle case reports to clinical trials. We were particularlyinterested in the reporting of pain intensity and tender mus-cle points before and after treatment.

Methods

Search strategy

To get a first overview, the following electronical databaseswere used to find articles on fibromyalgia and homeopa-thy: Medline, PubMed, Embase, AMED, CAMbase11 and thelibrary of the Carstens Foundation. The literature search,which was constructed around search terms for ‘home-opathy’ and ‘fibromyalgia syndrome’, was adapted andtranslated for each database, if necessary. For example, thefollowing search strategy was used on the PubMed/MEDLINEdatabase:

(Hoemopathy [MeSH Terms] OR homeopathy [Title/Abstract]) AND (fibromyalgia [MeSH Terms] OR fibromyalgia[Title/Abstract] OR fibrositis [Title/Abstract]).

In addition, an internet search was performed usingGoogle Scholar adding the search terms ‘‘study’’ and ‘‘casereport’’ to the above search terms. Finally, we also screenedthe database ‘‘Erfahrungsschatz Homöopathie’’ (ThiemePublishers) to find additional material. All articles found thisway were fully read and their reference lists were checkedfor further relevant publications. Articles included in thiscomprehensive review were classified with respect to yearof publication, research design, homeopathic treatment,patient demographics, the number of patients involved, andmain outcomes/results. The search was performed betweenOctober 2012 and February 2013.

It should be noted that fibrosis is the formation of excessfibrous connective tissue in an organ or tissue in a reparativeor reactive process. This can be a reactive, benign, or patho-logical state. Fibrositis, on the other hand, was historicallyused to name what is now labelled fibromyalgia.

Study selection

Inclusion/exclusion criteriaNon-randomized and randomized controlled trials, uncon-trolled observational trials, case studies, and case serieswere eligible when they assessed the effects of a homeo-pathic intervention on patient-related outcomes in patientswith fibromyalgia. There were no language restrictions.No restrictions in terms of age were applied (children,adolescents, adult and elderly were included). Stud-ies with patients with a serious concomitant medicalillness as an inclusion criterion were excluded. ‘Casereports’ also included case series of more than onecase.

Selection processTwo review authors independently screened abstracts iden-

tified during literature and read potentially eligible articlesin full to determine whether they met the eligibilitycriteria. No other study or report characteristics were pre-specified other than it having to be a clinical study of FMS

rgTd

733

atients who received a homeopathic treatment of someort.

tatistical analysis

eta-analysis on the effects of homeopathic treatment wasarried out according the established guidelines for non-andomized and randomized controlled trials.12,13 When arial was found to be eligible, data were extracted andntered into a pre-specified data sheet. If the number oftudies included in a meta-analysis is small, random effectsests are regarded as only approximate and fixed effectsests are regarded as more precise.41 Thus, a fixed effectsodel was used. Standardized mean differences (SMD) with

5%CI were calculated as the difference in means betweenroups divided by the pooled standard deviation and meta-nalyzed using the generic inverse variance method with

fixed effect model.13 Cohen’s d is defined as the differ-nce between two means divided by a standard deviationor the data. From this one can calculate the SMD. Effectizes can then be interpreted by keeping to pointers such as.2 ≡ small, 0.5 ≡ moderate and 0.8 ≡ large.42

Single case reports were exempt from meta-analysis aso SMD could be calculated.

Heterogeneity between trials was assessed by stan-ard Chi-Square-tests and the I2-coefficient measuring theercentage of total variation across studies due to true het-rogeneity rather than chance. Overall estimates of thereatment effect were obtained from fixed effects meta-nalysis. Results were displayed using a forest plot. Dueo the small number of eligible studies further analysis byeans of meta-regression was omitted.Subgroup analyses were conducted for type of homeo-

athic treatment (individualized homeopathy; homotoxicol-gy; potentized anthroposophic remedies).

uality assessment

uality of non-randomized and randomized controlled tri-ls was rated according to the Cochrane Quality Assessmentool for Quantitative Studies. Rating included (A) selectionias, (B) study design, (C) confounders, (D) blinding, (E) dataollection methods, (F) withdrawals and drop outs, (G) inter-ention integrity, and (H) appropriateness of analysis. Bothata extraction and quality assessment was cross-checkedy an independent rater. In case of disagreement consen-us between the raters was obtained by involving a thirdeviewer. The reporting of the results follows the establishedRISMA guidelines.30

esults

n total, we retrieved 10 case-reports, 3 uncontrolled obser-ational trials, 1 non-randomized controlled trial and 4

andomized controlled trials on homeopathy for fibromyal-ia from the literature search (for flow chart, see Fig. 1).hey are listed chronologically by means of their researchesign in Tables 1—3.

734 K. Boehm et al.

Potentially relevant studies addressing

fibromyalgia and ho meopat hy (n=1 64)

setacilpud–seidutsdedulcxE

(n=108)

Potentially appropriate studies to be

included (n=56)

Excluded st udies (n=48)

Included controlled studies; suitable for data extraction

(n=8)

Randomized and

single-blinded

(n=6)

Non-randomized

(n=2)

Included in meta-analysis

(n=6)

rt of

C

CTp

tN3

aLhbioa

wrcaf

ydLlp

wpda

fps

5rpTbh

UT(

gtbct

Fig. 1 Flow cha

haracteristics of included studies

ase reportsen case reports focussing the use of homeopathy foratients suffering on fibromyalgia were found (Table 1).

Gemmell et al.14 reported the outcomes of three patientsreated with Rhus toxicodendron 6 times for at least 21 days.o patient, however, showed sustained improvement after0 days.

Klein15 reported on the work with a male patient, age 54nd suffering from FMS with fatigue and pain. He receivedoxosceles reclusa (spider) 30c. After taking the first doseis back symptoms were increasingly worse for 7 days, thenetter but still problematic for 3 weeks. A more dramaticmprovement was observed after this point. The symptomsf fibromyalgia, pain and depression all improved remark-bly. The 30c dosage was repeated once.

Fleisher16 reported on a 56 year old business womanith severe chronic fatigue and fibromyalgia. After intensive

epertorising she was prescribed a single dose of Crotalusascavella 1 M. Three months later, she reported that fatiguend muscle pain were reduced by more than 80%. She alsoelt more relaxed.

Jones and Whitmarsh17 reported two cases of 64 and 45ear old females. While the first one received Rhus toxico-

endron 6c three times daily followed by Lachesis 200c andM1, the second one received Calcium carbonicum 200c fol-owed by LM1 and LM2. The first patient reported less muscleain as well as uplifted mood and a better ability to cope

nnFm

study inclusion.

ith problems. The second patient also reported decreasedain and increased calmness. She also felt strong enough toeal with her marital problems, which had been present for

long time.Saltzman18 prescribed Papaver somniferum 30c to a

emale patient with fibromyalgia and gastroparesis, thehysical and mental complaints disappeared and the patienthowed very positive personal and spiritual development.

Walters et al.19 also reported two cases out of a study of6 patients who had received individualized homeopathicemedies. One female patient reported no improvement ofhysical complaints, but she felt more positive and fulfilled.he other female patient showed improved general health,ut increased shoulder pain, which homeopathy could notelp with.

ncontrolled clinical trialshree observational studies were included in this reviewTable 2).

One Dutch trial20 included 42 patients attending theireneral practitioner for FMS complaints. Patients were askedo complete the Fibromyalgia Impact Questionnaire (FIQ) ataseline, after five and ten weeks of treatment. Treatmentonsisted of Hepar Magnesium D10 intravenously adminis-ered weekly for 10 weeks. After five weeks, the rating of

ine out of ten FIQ items demonstrated a statistically sig-ificant improvement. After ten weeks the rating of sevenIQ items demonstrated a statistically significant improve-ent. The results show that a large subgroup demonstrates

Homeopathy in the treatment of fibromyalgia 735

Table 1 Study characteristics: case reports.

Reference Patients (N, age,diagnosis)

Homeopathic remedy Results

Gemmell et al.,1991

69 year old female Rhus toxicodendron 6x, threetimes daily for at least 21days + spinal manipulations

Homeopathic remedy failed to improvepain.

48 year old female Rhus toxicodendron 6x, threetimes daily for at least 21 days

Homeopathic remedy failed to improvepain.

46 year old female Rhus toxicodendron 6x, threetimes daily for at least 21 days

Short term improvement of pain on anumerical rating scale was observed,afterwards return to baseline pain

Klein, 2001 54 year old male Loxosceles reclusa (spider) 30c Back pain worsened for 7 days followingthe first dose, and then improveddramatically. Patient also reportedstrong improvement of other symptoms,such as fatigue and depression.

Fleisher, 2004 56 year old female Single dose of Crotaluscascavella 1 M

Patient reported that fatigue and musclepain were reduced by more than 80%.She also felt more relaxed.

Jones andWhitmarsh, 2008

64 year old female Rhus toxicodendron 6c, threetimes daily; Lachesis200c,three times for a day,then LM1 five drops daily

Patient reported lifted mood, betterability to cope and less muscle pain.

45 year old female Calcium carbonicum 200c,three times for a day, then LM1five drops daily, later increaseto LM2

Patient reported decrease of pain andincreased calmness. She also decided todeal with marital problems.

Saltzman, 2008 40 year old female Papaver somniferum 30c Patient showed very positive personaldevelopment, physical and mentalcomplaints disappeared.

Walters, 2011 female, age notreported

Carcinosin, Aurum, Syphilinum,Nux vomica, Natrummuriaticum, Folliculinum

No improvements of physical symptomwere observed, but patients felt positiveand fulfilled.Diarrhoea was reported during one ofthe treatment phases.

female, age notreported

Rhus toxicodendron, Aurum,Carcinosum

Patient showed improved generalhealth, but she also reported shoulder

tlGl(omp

RIpowo

an improvement of more than 20% and a smaller group (34%after five weeks and 20% after ten weeks) that demonstratesan improvement of more than 30% (up to 81%) of the totalFIQ-score.

Another uncontrolled trial by Walters et al.19 found that29 patients, who had received individual homeopathic reme-dies, showed significant improvement in the FibromyalgiaImpact questionnaire, symptoms measured by the MYMOPand quality of life. The homeopathic treatment included amaximum of 9 homeopathic consulting sessions; outcomeswere assessed after 52 weeks.

The same author published a service evaluation of 56patients with fibromyalgia;21 however no cumulative resultswere reported in the manuscript. The 2 cases described inthis study can be found in Table 1.

Controlled clinical trialOne controlled but not randomized clinical trial fromSpain was found,22 see Table 3. Researchers investigated

natb

pain and exotosis.

he short term effectiveness (8 weeks) of a pharmaco-ogical antihomotoxic treatment (Traumeel®, Spascupreel®,raphites Homaccord®, Cerebrum compositum® and Tha-

amus compositum®) versus a placebo (physiologic serum)double blind) in 20 patients diagnosed of fibromyalgia. Thebtained results showed a significant improvement in theuscular and psychological symptoms with regard to therevious state in the treatment groups.

andomized controlled trialsn a British study by Fisher23 for 3 months 24 patients wererescribed, according to indication, one of three home-pathic remedies (Arnica, Bryonia, Rhus toxicodendron)here each patient remaining on the same remedy through-ut. They were followed monthly on the parameters pain,

umber of tender spots and sleep. An ‘indication score’ wasllotted to each prescription. The results showed a statis-ically significant improvement for the homeopathy group,ut only when the prescribed remedy was well indicated.

736 K. Boehm et al.

Table 2 Study characteristics: uncontrolled observational trials.

Reference Patients (N, age, diagnosis) Homeopathic remedy Results

Baars, 2010 42 patients withfibromyalgia diagnosed byrheumatologistAge: 47 years (range 22—65)Females: 39/42

Hepar Magnesium 10dintravenouslyWeekly administration of10 ml for 10 weeksNo adjunctive therapies

Significant improvement in theFibromyalgia impact questionnaire(FIQ) at week 10

Walters et al.,2011

29 patients withfibromyalgiaGender and age unknown

Individualized homeopathicremediesMax. 9 homeopathicconsulting sessions, 1 heach, at 5 week intervals

Significant decrease in Fibromyalgiaimpact questionnaire (FIQ) at week52.Significant decrease in measure yourmedical outcomes profile at week 52(MYMOP-symptoms).Significant increase in quality of lifeat week 40 (EuroQol).

Walters, 2011 56 patients withfibromyalgiaGender and age unknown

Individualized homeopathicremedies

No cumulative result reported(results of two cases see table)

dpfir1titatp

n(ppqmsnttp

tpts(grFatt

M

MoPPPsd

vi−Irnl

Q

TwmWpbo

D

Tadgated so far, including a variety of complementary therapies.Perry et al.8 had reported on the effects of homeopathy in

In 1988 Fisher24 published another study where they ran-omized 30 patients suffering from fibromyalgia. The activereparation was Rhus toxicodendron 6c (Boiron) preparedrom a tincture of the leaves of poison oak diluted 1:99n ethanol and then vigorously shaken. This process wasepeated six times to give the 6c potency — a dilution of02 of the tincture. This was then put up on 125 mg lac-ose tablets. Results showed that the patients did bettern all variables when they took active treatment ratherhan placebo. The number of tender spots was reduced bybout a quarter (P < 0005). Thus, it was concluded that Rhusoxicodendron 6c was effective for a selected subgroup ofatients.

The American research team Bell et al.25 randomized = 62 fibromyalgia patients to receive oral daily liquid LM1/50 000) potencies with an individually chosen homeo-athic remedy or an indistinguishable placebo. Outcomearameters included tender point count, tender point pain,uality of life, pain, mood and global health and they wereeasured at baseline, 2 months and 4 months. Results

howed that participants on active treatment showed sig-ificantly greater improvements in tender point count andender point pain, quality of life, global health and arend towards less depression compared with those onlacebo.

Finally, in 2009 another British research group publishedhe findings of an RCT led by Relton et al.26 Forty sevenatients were recruited of which 11 had dropped-out byhe end of the trial. Adjusted for baseline, there was aignificantly greater mean reduction in the FIQ total scorefunction) in the homeopath care group than the usual careroup (−7.62 versus 3.63). There were significantly greatereductions in the homeopathy group in the McGill pain score,IQ fatigue and tiredness upon waking scores. Researcherslso found a small effect on pain score (0.21, 95%CI −1.42o 1.84); but a large effect on function (0.81, 95%CI −8.17

o 9.79).

Data from all RCTs is presented in Table 3.fit

eta-analysis

eta-analyses of RCTs revealed effects of homeopathyn tender point count (SMD = −0.42; 95%CI −0.78, −0.05;

= 0.03), pain intensity (SMD = −0.54; 95%CI −0.97, −0.10; = 0.02), and fatigue (SMD = −0.47; 95%CI −0.90, −0.05; = 0.03) compared to placebo. Pain on the McGill pain sen-ory or affective pain subscales, and depression did notiffer between groups (Fig. 2).

In subgroup analyses, when only studies that used indi-idualized homeopathy were considered, the effect on painntensity was no longer significant (SMD: −0.36; 95%CI:0.85 to 0.13; P = 0.15). Heterogeneity was reduced from

2 = 42% (P = 0.18) to I2 = 13% (P = 0.28). No other changes inesults were found. Due to the paucity of included studies,o separate subgroup analyses for studies on homotoxico-ogical or anthroposophic treatment were possible.

uality assessment

wo randomized trials had low risk of selection bias,25,26

hile the two randomized cross-over trials did not reportethods of randomization or allocation concealment.23,24

hile only 2 trials reported adequate blinding of partici-ants and personnel,24,25 all trials but 122 reported adequatelinding of outcome assessment. Risk of attrition, reportingr other bias was low in most trials.

iscussion

he treatment of fibromyalgia is still a challenge for patientsnd physicians. Due to quite heterogeneous courses of theisease a variety of therapeutic options have been investi-

bromyalgia and concluded that ‘‘homoeopathy was betterhan the control interventions in alleviating the symptoms

Hom

eopathy in

the treatm

ent of

fibromyalgia

737

Table 3 Study characteristics: controlled and randomized controlled trials.

Reference Study type Patients (N, age,diagnosis)

Intervention groups (programmelength, frequency, duration)

Follow-up Outcome measures Results

Treatment Control

Egocheagaand del Valle,2004

CCT (casecontrol study)

20 patients withfibromyalgia

Pharmacologicalantihomotoxic injection(Traumeel, Spascupreel,Graphites Homaccord,Cerebrum compositum,Thalamus compositum)8 weeks, twice weekly

Placebo injection8 weeks, twiceweekly

8 weeks (1) Pain intensity on anumeric rating scale(NRS)

(1) Significant groupdifference in favour ofhomeopathy

Gender: female onlyAge: 28—64 years

(2) Psychologicalwellbeing on anumeric rating scale(NRS)

(2) No significant groupdifferences

Fisher, 1986 RCT 24 patients withfibrositis(fibromyalgia) andindication for RhustoxicodendronprescriptionGender and ageunknown

One of the three remedies:Arnica montana, Bryoniaalba or Rhus toxicodendron(potency 6c)3 months, twice daily intake

Placebo3 months, twicedaily intake

3 months (1) Tender pointcounts (TPC)(2) Pain intensity on avisual analogue scale(VAS)(3) Analgesicconsumption

(1) No significant groupdifferences at 3 months(2) No significant groupdifferences at 3 months(3) Not reported

Fisher, 1988 RCT 30 patients withfibromyalgiaMean age: 48.4 years(range from 29 to 64)Females: 23/30

Rhus toxicodendron(potency 6c)1 month, three times daily2 tablets

Placebo1 month, threetimes daily 2tablets

1 month (1) Tender pointcounts (TPC)(2) Pain intensity on avisual analogue scale(VAS)(3) Sleep quality on avisual analogue scale(VAS)

(1) No significant groupdifferences at 1 month,significant groupdifferences regardingchange from baseline infavour of homeopathy(2) No significant groupdifferences at 1 month,significant groupdifferences regardingchange from baseline infavour of homeopathy(3) No significant groupdifferences at 1 month

738

K. Boehm

et al.

Table 3 (Continued)

Reference Study type Patients (N, age,diagnosis)

Intervention groups (programmelength, frequency, duration)

Follow-up Outcome measures Results

Treatment Control

Bell et al.,2004

RCT 62 patients withfibromyalgiaaccording to the ACRcriteriaMean age: 49.1 ± 9.9years (Homeopathy)47.9 ± 10.8 years(Placebo)Females:29/30 (Homeopathy)29/32 (Placebo)

Individualized homeopathicremedies, daily intakeHomeopathic consulting at0, 2, 4, 6 months

Placebo, dailyintakeHomeopathicconsulting at 0, 2,4, 6 months

3 months (1) Tender pointcount (TPC)(2) McGill painquestionnaire shortform (sensorycomponent, affectivecomponent)(3) Profile of Moodstates (POMS)(4) Appraisal ofdisease interactionwith life goals(5) Global healthrating

(1) Significant groupdifference in favour ofhomeopathy(2) No significant groupdifferences(3) No significant groupdifferences(4) Significant groupdifference in favour ofhomeopathy(5) Significant groupdifference in favour ofhomeopathy

Relton et al.,2009

RCT 47 patients withfibromyalgiaaccording to the ACRcriteriaMean age: 43.9 ± 8.9years (Homeopathy)47.4 ± 9.2 years(Usual care)Females: 22/23(Homeopathy)22/24 (Usual care)

Individualized homeopathictreatment1 initial homeopathicinterview of 60 min4 homeopathic interviewsof 30 min every 4—6 weeks+Usual care

Usual care 22 weeks (1) Tender pointcount (TPC)(2) McGill painquestionnaire(sensory component,affective component)(3) Fibromyalgiaimpact questionnaire(FIQ)(4) Hospital Anxietyand Depression Scale(HADS)(5) Measure yourmedical outcomesprofile (MYMOP)(6) Pain intensity on avisual analogue scale(VAS)(7) Quality of life(EuroQol)

(1) No significant groupdifferences at 22 weeks(2) No significant groupdifferences at 22 weeks,significant groupdifferences regardingchange from baseline infavour of homeopathy(3) No significant groupdifferences at 22 weeks,significant groupdifferences regardingchange from baseline infavour of homeopathy(4) No significant groupdifferences at 22 weeks(5) No significant groupdifferences at 22 weeks(6) No significant groupdifferences at 22 weeks,significant groupdifferences regardingchange from baseline infavour of homeopathy(7) No significant groupdifferences at 22 weeks

Homeopathy in the treatment of fibromyalgia 739

Fig. 2 Forrest plots for outcomes.

7

oc

ficaAfnt

oioiabfict

M

IpttbtsEs

S

AFWott

paarolWd

awsdtbvMm

S

Ttawwarp‘b

hatti‘Ifgettite

sccmLeirp

mrdoieacwttaopr

it

heterogeneity of interventions, which in homeopathy is aquite common meta-analytic issue due to the individualistic

40

f FMS’’ but also pointed out problems in the quality of thelinical trials.

In 2010 the American College of Rheumatology criteriaor the diagnosis of FMS changed and all the RCTs includedn this review were conducted before then. The most signifi-ant difference is that the 2010 guidelines no longer require

tender point count for the diagnosis of FMS. The previousCR criteria proposed for the classification of fibromyalgiaor the included studies were (1) widespread pain in combi-ation with (2) tenderness at 11 or more of the 18 specificender point sites.31

Our analysis for the first time provides a comprehensiveverview of homeopathic literature in the treatment of FMSncluding a summary of single case descriptions, a reviewf current observational studies and a meta-analysis of RCTsncluding 1 RCTs not included in prior reviews. Thus, ourpproach not only provides the most current evidence basey quantifying the effectiveness homeopathic treatment ofbromyalgia but also covers the perspective of the practi-al homeopath, and delivers a source of information i.e. inerms of single case descriptions.

eta-analysis

n accordance to Perry et al.8 quality assessment revealedotential sources of bias in the included studies. Similarlyo another meta-analysis on the use of Hypericum perfora-um (St. John’s Wort) for pain conditions in dental practicey Raak et al.27 we found that studies had several limita-ions by means of biases. In contrast to Raak et al.27 oldertudies like those of Fisher23,24 had a sufficient quality whilegocheaga and del Valle22 i.e. failed on reporting severaltudy essentials.

ingle cases

ll single cases were published after the studies ofisher.23,24 Thus, descriptions of Gemmell et al.,14 Jones andhitmarsh,17 and Walters21 more or less adopted the home-

pathic treatment strategy described there and used Rhusoxicodendron as a single remedy14 or with some modifica-ions in Jones and Whitmarsh17 and Walters.21

Rhus toxicodendron according to the repertory is appro-riate in restless patients with apprehension at nightccompanied by heavy feeling in the head, joint stiffnessnd a sensibility for cold weather. However there are otheremedies like Calcium carbonicum or the exotic L. reclusar P. somniferum which potentiated in 30c also seem toower FMS symptoms in the cases described by Jones andhitmarsh,17 Klein15 and Saltzman.18 However Rhus toxico-

endron still remains the most evaluated remedy so far.Our review also includes potentiated remedies which

ccording to their origin cannot directly be interconnectedith the classical homeopathic approach of a potentiated

ingle agent. For example the study of Egocheaga andel Valle22 used pharmacological antihomotoxic injection inheir RCT and thus not only differ in the type of remedyut also in its application. This also holds for the obser-ational study of Baars and Ellis20 applying 10 ml of Hepar

agnesium 10d, a potentiated remedy of anthroposophicaledicine intravenously.

ns

K. Boehm et al.

trengths and limitations

his is the first available meta-analysis on homeopathy forreating fibromyalgia. In order to evaluate the totality ofvailable evidence, uncontrolled and single case studiesere included besides randomized trials. It is debatablehether this meta-analysis is limited by the inclusion of tri-ls of different study design. According to recent discussionegarding the use of hierarchical research evidence we pro-ose that a circular evidence perspective, specifically anevidence mosaic’ can be adapted, as previously suggestedy academic proponents of CAM research.34,39,40

Additionally, one could argue that the data presentedere are too heterogeneous for carrying out a meta-analysisnd that this should only be considered when a group ofrials is sufficiently homogeneous in terms of conditionreated, interventions and outcomes.35 Up to date theres no guideline to explain the meaning of ‘‘sufficiently’’ or‘meaningful’’ in this context except that Higgins mentions2 values beyond 50% as a substantial heterogeneity.36 Inact, Higgins himself argues that ‘‘any amount of hetero-eneity is acceptable, providing both that the predefinedligibility criteria for the meta-analysis are sound and thathe data are correct. The challenge is then to decide onhe most appropriate way to analyze heterogeneous stud-es, and this will depend on the aims of the synthesis and,o an extent, the observed directions and magnitudes offfects.’’37

Furthermore, we know from previous research that alltatistical tests for heterogeneity are weak, including I2. Thelinical implications of this must be examined on a case byase basis. Putting too much trust in homogeneity of effectsay give a false sense of reassurance that one size fits all.

ack of evidence of heterogeneity cannot be interpreted asvidence of homogeneity. Conversely, putting too much trustn the presence of heterogeneity of effects may lead to spu-ious subgroup and exploratory analyses. Given that I2 is notrecise, 95% confidence intervals should always be given.38

A substantial limitation of this meta-analysis is the lowethodological quality of the included studies. As in prior

eviews,8 the interpretation of the findings is clearly limitedue to the insufficient reporting of research methodol-gy. Moreover, the heterogeneity of interventions in thencluded studies further limits interpretability. While Bellt al.25 and Relton et al.26 used a strictly individualizedpproach without restriction of the chosen remedy, otherontrolled trials (semi-)standardized the remedies thatere administered.23,24 Finally, while all other controlled

rials used single remedies based on classical homeopa-hy, Egocheaga and del Valle22 prescribed pharmacologicalntihomotoxic injection, i.e. complex remedies. The trialf Fisher28 did not report sufficient data in the originalublication; data for meta-analysis was extracted from aeanalysis.29

Furthermore, endpoint definitions differed among stud-es (see the above discussion on change of diagnosis ofender points in 2010).

We carried out a sensitivity analysis to counteract the

ature of the intervention. With the sensitivity analy-is heterogeneity was reduced although there remained a

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

Homeopathy in the treatment of fibromyalgia

substantial residual heterogeneity. However, the effect onpain intensity was no longer significant. Lack of patient leveldata prevented a further analysis of the current results strat-ified by age or sex.

Generally, we feel that our group of trials is sufficientlyhomogeneous in terms of condition treated, interventionsand outcomes.

Implications for further research

In line with prior reviews,8 the interpretability of evidencefound in this meta-analysis is limited by the low method-ological quality of the included studies. Future trials shouldensure rigorous methodology and reporting, mainly ade-quate randomization, allocation concealment, intention-totreat analysis, and blinding. Future research should alsotry to investigate the comparative effectiveness of classicalhomeopathy, clinical homeopathy, and the use of complexremedies.

Conclusions

Given the low number and included trials and the lowmethodological quality, any conclusion based on the resultsof this review have to be regarded as preliminary. However,as single case studies and clinical trials indicate a positiveeffect, homeopathy could be considered a complementarytreatment for patients with fibromyalgia.

Conflict of interest statement

The authors declare that they have no conflicts of interest.

Funding

No funding was received from any source by any of theauthors specifically for this review.

Box 1 Key messages• Available data suggest that homeopathic treatment

of fibromyalgia is helpful in pain management andreduction of fatigue.

• Further studies are merited in terms of multicentertrials to broaden the evidence base.

References

1. Häuser W, Hayo S, Biewer W, Gesmann M, Kühn-Becker H, Pet-zke F, et al. Diagnosis of fibromyalgia syndrome—–a comparisonof Association of the Medical Scientific Societies in Germany,survey, and American College of Rheumatology criteria. Clin JPain 2010;26(6):505—11.

2. Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Katz RS,Mease P, et al. The American College of Rheumatology prelim-inary diagnostic criteria for fibromyalgia and measurement ofsymptom severity. Arthritis Care Res 2010;62(5):600—10.

3. Wolfe F, Brahler E, Hinz A, Häuser W. Fibromyalgia preva-lence, somatic symptom reporting, and the dimensionality ofpolysymptomatic distress: results from a survey of the generalpopulation. Arthritis Care Res 2013;65(5):777—85.

2

741

4. Branco JC, Bannwarth B, Failde I, Abello Carbonell J, Blot-man F, Spaeth M, et al. Prevalence of fibromyalgia: asurvey in five European countries. Semin Arthritis Rheum2010;39(6):448—53.

5. Berger A, Dukes E, Martin S, Edelsberg J, Oster G. Charac-teristics and healthcare costs of patients with fibromyalgiasyndrome. Int J Clin Pract 2007;61(9):1498—508.

6. Berger A, Sadosky A, Dukes E, Martin S, Edelsberg J, OsterG. Characteristics and patterns of healthcare utilization ofpatients with fibromyalgia in general practitioner settings inGermany. Curr Med Res Opin 2008;24(9):2489—99.

7. Wahner-Roedler DL, Elkin PL, Vincent A, Thompson JM, OhTH, Loehrer LL, et al. Use of complementary and alterna-tive medical therapies by patients referred to a fibromyalgiatreatment program at a tertiary care center. Mayo Clin Proc2005;80(1):55—60.

8. Perry R, Terry R, Ernst E. A systematic review of homoeopa-thy for the treatment of fibromyalgia. Clin Rheumatol2010;29(5):457—64.

9. Langhorst J, Häuser W, Bernardy K, Lucius H, Settan M, Winkel-mann A, et al. Complementary and alternative therapies forfibromyalgia syndrome. Systematic review, meta-analysis andguideline. Schmerz 2012;26(3):311—7 [in German].

0. Baranowsky J, Klose P, Musial F, Häuser W, Dobos G, LanghorstJ. Qualitative systemic review of randomized controlled tri-als on complementary and alternative medicine treatments infibromyalgia. Rheumatol Int 2009;30(1):1—21.

1. Ostermann T, Zillmann H, Matthiessen PF. CAMbase–the real-isation of an XML-based bibliographical database system forcomplementary and alternative medicine. Z ärztl Fortbild QualSich 2004;98(6):501—7 [in German].

2. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Introduc-tion to meta-analysis (statistics in practice). West Sussex: JohnWiley & Sons; 2009.

3. Higgins JPT, Green S. Cochrane handbook for systematic reviewsof interventions. West Sussex: John Wiley & Sons Ltd.; 2008.

4. Gemmell HA, Jacobson BH, Banfield K. Homeopathic Rhus tox-icodendron in the treatment of fibromyalgia. Chiropr J Aust1991;21(1):2—6.

5. Klein L. Loxosceles reclusa, a new proving. Homeopathic2001;21(3):26—9.

6. Fleisher MA. A case of chronic fatigue and fibromyalgia. Am JHomeopath Med 2004;97(2):109—10.

7. Jones A, Whitmarsh T. Fibromyalgia. Health Homeopathy2008;(Summer):21—3.

8. Saltzman S. A case of severe fibromyalgia and gastroparesis. AmJ Homeopath Med 2008;101(4):227—8.

9. Walters C, Raw J, Smith C, Adebajo A. Fibromyalgia syndrome.A service evaluation. Homeopathic 2011;30(3):9—11.

0. Baars EW, Ellis EL. The effect of Hepar Magnesium D 10on fibromyalgia syndrome: a pilot study. Eur J Integr Med2010;2(1):15—21.

1. Walters C. Don’t touch me, I hurt too much. Homeopathic2011;29(4):23—6.

2. Egocheaga J, del Valle M. Use of antihomotoxic drug therapy forthe management of symptoms associated to fibromyalgia. RevSoc Esp Dolor 2004;11:4—8 [in Spanish].

3. Fisher P. An experimental double-blind clinical trial method inhomoeopathy. Br Homeopath J 1986;75(3):142—7.

4. Fisher P. Rhus toxicodendron in the treatment of Fibromyalgia:a double-blind, placebo controlled trial, with cross over. J OMHJ1988;1(3):26—8.

5. Bell IR, Lewis 2nd DA, Brooks AJ, Schwartz GE, Lewis SE,Walsh BT, et al. Improved clinical status in fibromyalgia

patients treated with individualized homeopathic remediesversus placebo. Rheumatology 2004;43(5):577—82.

6. Relton C, Smith C, Raw J, Walters C, Adebajo AO, Thomas KJ,et al. Healthcare provided by a homeopath as an adjunct to

7

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

42

usual care for Fibromyalgia (FMS): results of a pilot RandomisedControlled Trial. Homeopathy 2009;98(2):77—82.

7. Raak C, Bussing A, Gassmann G, Boehm K, Ostermann T. Asystematic review and meta-analysis on the use of Hypericumperforatum (St. John’s Wort) for pain conditions in dental prac-tice. Homeopathy 2012;101(4):204—10.

8. Fisher P, Greenwood A, Huskisson EC, Turner P, Belon P. Effectof homeopathic treatment on fibrositis (primary fibromyalgia).BMJ 1989;299(6695):365—6.

9. Colquhoun D. Re-analysis of clinical trial of homoeopathic treat-ment in fibrositis. Lancet 1990;336(8712):441—2.

0. Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group.Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6(6):e1000097,http://dx.doi.org/10.1371/journal.pmed1000097.

1. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C,Goldenberg DL, et al. The American College of Rheuma-tology 1990 criteria for the classification of fibromyalgia.Report of the multicenter criteria committee. Arthritis Rheum1990;33(2):160—72.

2. Carville SF, Arendt-Nielsen S, Bliddal H, Blotman F, Branco JC,Buskila D, et al. EULAR evidence-based recommendations forthe management of fibromyalgia syndrome. Ann Rheum Dis2008;67(4):536—41.

3. Sarzi-Puttini P, Atzeni F, Salaffi F, Cazzola M, Benucci M, MeasePJ. Multidisciplinary approach to fibromyalgia: what is theteaching? Best Pract Res Clin Rheumatol 2011;25(2):311—9,http://dx.doi.org/10.1016/j.berh.2011.03.001.

4

K. Boehm et al.

4. Walach H. Zirkulär statt hierarchisch: Methodologische Grund-prinzipien bei der Evaluation der therapeutischen Effektevon Komplementärmedizin und anderer komplexer Maßnahmen[Circular instead of hierarchical: Methodological principles inthe evaluation of complementary medicine and other com-plex interventions] Informatik. Biomet Epidemiol Medizin Biol2004;35:229—42.

5. Higgins JP, Thompson SG. Cochrane handbook for systematicreviews of interventions, Version 5.1.0 [updated March 2011].The Cochrane Collaboration; 2011.

6. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuringinconsistency in meta-analyses. Br Med J 2003;327:557e60.

7. Higgins JPT. Commentary heterogeneity in meta-analysis shouldbe expected and appropriately quantified. Int J Epidemiol2008;37(5):1158—60.

8. Ioannidis JPA, Patsopoulos NA, Evangelou E. Uncertainty in het-erogeneity estimates in meta-analyses. BMJ 2007;335:914.

9. Walach H, Falkenberg T, Fønnebø V, Lewith G, Jonas WB. Cir-cular instead of hierarchical: methodological principles for theevaluation of complex interventions. BMC Med Res Methodol2006;6.

0. van Haselen R. Medical study formats: an overview. J BiomedTher 2010;4(2):26—7.

1. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Fixed-

effect versus random-effects models. In: Introduction tometa-analysis. Chichester, UK: John Wiley & Sons, Ltd; 2009.

2. Cohen J. Statistical power analysis for the behavioral sciences.Hillsdale, MI: Lawrence Erlbaum Associates; 1988.