Sys ema ic Re iew
The mos e ec i e he apeu ic exe cises o pain in ensi y in women wi h
ib omyalgia: A sys ema ic e iew and ne wo k me a-analysis
☆
´
Al a o-Jos´
e Rod íguez-Domínguez
a
, Manuel Rebollo-Salas
b,*
, Raquel Chill´
on-Ma ínez
a
,
Abel Rosales-T is ancho
c
, Inmaculada Villa-del-Pino
b
, Jos´
e-Jesús Jim´
enez-Rejano
b
a
Depa men o Heal h and Spo s, Pablo de Ola ide Uni e si y, Se ille, Spain
b
Depa men o Physical The apy, Uni e si y o Se ille, Se ille, Spain
c
Depa men o Applied Economics I, Uni e si y o Se ille, Se ille, Spain
ARTICLE INFO
Keywo ds:
Fib omyalgia
Pain
Physical he apy
Rehabili a ion
The apeu ic exe cise
ABSTRACT
Backg ound: The apeu ic exe cise (TE) is he only in e en ion wi h s ong ecommenda ion o ib omyalgia.
Howe e , he e is con o e sy as o which ype o exe cise is he mos bene icial.
Objec i e: To de e mine which TE app oach is he mos e ec i e in educing pain in ensi y in women wi h
ib omyalgia.
Me hods: A sys ema ic e iew was pe o med wi h a ne wo k me a-analysis (NMA). Six da abases we e sea ched
om incep ion un il Janua y 2024. Randomized con olled ials (RCTs) e alua ing he e ec s o TE on pain
in ensi y we e included in women wi h ib omyalgia. Me hodological quali y was assessed using he Physio-
he apy E idence Da abase scale. The size o he e ec and he clinically impo an di e ence we e de e mined in
he sho - e m (≤3 mon hs) and long- e m (>3 mon hs).
Resul s: Six y-one s udies we e iden i ied, o which 51 we e included in he quan i a i e syn hesis (n =2873).
Fi een TE in e en ions and eigh compa ison in e en ions (compa a o s) we e iden i ied. Aqua ic exe cise (p-
sco e: 0.8713) was ound o p o ide bes bene i s in he sho - e m and esis ance aining in he long- e m (p-
sco e: 0.9749). S a is ically signi ican di e ences we e ound in a o o aqua ic exe cise, Pila es, qigong,
esis ance aining, i ual eali y, mixed exe cise, and ae obic exe cise (in he sho - e m) and in a o o
esis ance aining, dance, unc ional aining, aqua ic exe cise, i ual eali y, and ae obic exe cise (in he long-
e m) compa ed o usual ca e.
Conclusion: Wi h a mode a e le el o e idence, ou NMA shows ha , in he sho - e m, aqua ic exe cise is he
mos e ec i e TE in e en ion o educe pain in ensi y in women wi h ib omyalgia, while esis ance aining is
he mos e ec i e in he long- e m. Mo e RCTs a e needed o s eng hen hese indings.
In oduc ion
Fib omyalgia is a ch onic synd ome cha ac e ized by widesp ead
musculoskele al pain, ch onic a igue, sleep dis u bances, and physical
disabili y.
1–4
I is p e alen in 2 o 4 % o he wo ld’s popula ion,
2–6
a ec ing mainly women.
5,7,8
Al hough i is hypo hesized ha ib omy-
algia is due o a cen al sensi iza ion p ocess, he e iopa hological
mechanisms a e s ill unclea .
3,9–12
S udies ha e iden i ied se e al mus-
cle abno mali ies, such as a educ ion in ype II ibe s and al e a ions in
muscle con ac ion and me abolism, which can con ibu e o a igue and
pain in pa ien s wi h ib omyalgia.
13,14
These indings may explain he
bene i s o physical exe cise o hese pa ien s.
Physical exe cise is he only ea men wi h a s ong ecommenda-
ion o ib omyalgia.
2,3
Se e al e iews ha e e alua ed he e ec s o
di e en he apeu ic exe cise (TE) app oaches in pa ien s wi h ib o-
myalgia, wi h bene i s obse ed in almos all o hem.
15–22
Howe e ,
he e is con o e sy as o which app oach is he mos bene icial. The
Coch ane e iew
23
o 2007 classi ied ae obic exe cise as he "gold
s anda d" o he ea men o ib omyalgia, al hough in hei 2017 e-
iew,
18
hey concluded ha his exe cise app oach may ha e li le o no
long- e m e ec on pain.
To da e, we ha e no ound any publica ions ha compa es all TE
☆
Regis a ion numbe : CRD42021293015.
* Co esponding au ho a : Depa men o Physical The apy, Uni e si y o Se ille, C/ A icena s/n, Se ille 41009, Spain.
E-mail add ess: [email p o ec ed] (M. Rebollo-Salas).
Con en s lis s a ailable a ScienceDi ec
B azilian Jou nal o Physical The apy
jou nal homepage: www.else ie .com/loca e/bjp
h ps://doi.o g/10.1016/j.bjp .2025.101226
Recei ed 23 Ap il 2024; Recei ed in e ised o m 9 Augus 2024; Accep ed 28 Ap il 2025
B az J Phys The 29 (2025) 101226
A ailable online 3 May 2025
1413-3555/© 2025 The Au ho s. Published by Else ie España, S.L.U. on behal o Associação B asilei a de Pesquisa e Pós-G aduação em Fisio e apia. This is an
open access a icle unde he CC BY license ( h p://c ea i ecommons.o g/licenses/by/4.0/ ).
app oaches wi h each o he o he ea men o pain in ib omyalgia.
Because he disease a ec s mainly women and he esponse o ea men
is in luenced by a mul i ude o ac o s, including gende , he objec i e o
his s udy was o analyze he e icacy o di e en TE app oaches in pain
in ensi y in women wi h ib omyalgia and o iden i y which exe cise
app oach is he mos e ec i e h ough a ne wo k me a-analysis (NMA)
o andomized clinical ials (RCTs).
Me hods
A sys ema ic e iew wi h an NMA ollowing he PRISMA-NMA
guidelines.
24
The p o ocol was egis e ed in he PROSPERO da abase.
Ou sea ch was conduc ed om da abase incep ion un il Janua y 14,
2024. The da abases selec ed we e MEDLINE, CENTRAL, Embase, Web
o Science (WoS), Cumula i e Index o Nu sing and Allied Heal h
Li e a u e (CINAHL), and Scopus. The sea ch s a egy and keywo ds
used a e gi en in Supplemen a y ma e ial (Table S.1)
To minimize publica ion bias, a sea ch was pe o med on Clin-
icalT ials.go . Addi ional eco ds we e sea ched by hand om ele an
li e a u e e iews o supplemen he indings o he da abase.
Eligibili y c i e ia
The inclusion c i e ia ollowed he PICOS (pa icipan s, in-
e en ions, compa a o s, ou comes, s udy design) s a egy:
1. Type o s udy: RCTs.
2. Pa icipan s: adul women diagnosed wi h ib omyalgia acco ding o
he c i e ia o he Ame ican College o Rheuma ology o ib omy-
algia (ACR 1990/2010/2016).
3. In e en ions: s udies ha included any o m o TE as he only
in e en ion o a combina ion o exe cises in any o he g oups we e
selec ed.
4. Compa a o s: any ea men
5. Ou comes: pain in ensi y, e alua ed wi h isual analog scale (VAS
0–10, 0–100)
T ials published in languages o he han English, Spanish, F ench,
I alian, o Po uguese we e excluded, as we e ials ha combined ET
wi h o he ea men s and ials ha included men.
Da a collec ion p ocess
Two e iew au ho s (AR-D, JJ-R) independen ly pe o med s udy
selec ion and da a ex ac ion. A hi d au ho (MR-S) was consul ed in
case o disag eemen . S udies ha did no epo he da a equi ed o
he me a-analysis we e excluded om he quan i a i e syn hesis. In
gene al, he da a collec ed we e he mean and s anda d de ia ion o
each ea men and pe iod s udied. Gi en he exis ence o s udies ha
p o ided o he ypes o measu emen s (such as median o qua iles Q1
and Q3), p io es ima es we e made o ob ain an app oxima ion o he
mean and s anda d de ia ion om he da a.
25,26
A s anda dized o m was used o da a ex ac ion, add essing pa -
icipan s, diagnos ic c i e ia, ype o in e en ion, ollow-up ime, and
esul s ob ained (Supplemen a y ma e ial -Table S.2).
Risk o bias
The me hodological quali y was e alua ed using he PED o scale.
27
The quali y o he s udies was e iewed by wo independen e alua o s
(IV-P, RC-M), and a hi d e alua o (AR-T) was consul ed when dis-
c epancies appea ed. The included s udies we e classi ied acco ding o
sco es o 9 o 10, 6 o 8, and ≤5 on he PED o scale and we e in e p e ed
as excellen , good, and ai quali y, espec i ely.
28
Ce ain y o he e idence
The CINeMA web applica ion was used o assess con idence in
indings om p ima y NMA. The CINeMA amewo k conside s six do-
mains ha a ec he le el o con idence in he NMA esul s: (a) wi hin-
s udy bias, (b) epo ing bias, (c) indi ec ness, (d) imp ecision, (e) he -
e ogenei y, and ( ) incohe ence. The e iewe s assessed he le el o
conce n o each ela i e ea men e ec o NMA as gi ing ise o "no
conce ns", "some conce ns", o "majo conce ns" in each o he six do-
mains. Then, judgmen s ac oss he domains a e summa ized in o a single
con idence a ing ("high", "mode a e", "low", o " e y low").
29
(Supple-
men a y ma e ial -Table S.3)
Clinically impo an di e ences (CID)
Acco ding o Coch ane,
16
he Ini ia i e on Me hods, Measu emen
and Pain Assessmen in Clinical T ials (IMMPACT) ecommended he
ollowing benchma ks o in e p e ing changes in pain in ensi y on a
nume ical a ing scale 0 o 10 in ch onic pain clinical ials: a) a 10 o 20
% dec ease is minimally impo an , b) a dec ease g ea e han 30 % is
mode a ely impo an , and c) a dec ease g ea e han 50 % is subs an-
ial.
30,31
The e o e, he minimum CID in ou come was in e p e ed as a
pain di e ence o 15 poin s ou o 100 (VAS 0–100). The CID was
de e mined om he di e ence ob ained o each in e en ion
compa ed o usual ca e.
Summa y measu es
The analysis included a quali a i e syn hesis (Supplemen a y ma e-
ial -Table S.2) and a quan i a i e syn hesis (pai wise me a-analysis and
NMA). VAS is epo ed as a con inuous measu e om 0 o 10 and om
0 o 100. The ou come da a was s anda dized om 0–100, using he
mean di e ence (MD) o pe o m he p ima y analyzes. A empo al di-
ision was es ablished in o wo pe iods: sho - e m (≤3 mon hs) and
long- e m (>3 mon hs). Fo he sho e m, only ollow-up esul s close
o 3 mon hs a e he end o he in e en ion we e selec ed. Fo he long
e m, only ollow-up esul s close o six mon hs a e he end o he
in e en ion we e selec ed.
In hose me a-analyses, a andom-e ec s model was used whe e
he e ogenei y be ween s udies was obse ed, while a ixed-e ec s model
was used o hose whe e homogenei y was obse ed. As indica i e o
homogenei y, I
2
coe icien alues <50 % and/o Chi
2
es alues o
homogenei y wi h p >0.05 we e aken. In all cases, he co esponding
o es plo is p esen ed (Supplemen a y ma e ial -Figs. 2–14). A pai wise
me a-analysis was pe o med using Re iew Manage so wa e e sion
5.4.
Rega ding NMA, he means p o ided by each s udy we e aken o
each o he ea men s pe o med, as well as he s anda d de ia ion o
each one, which allowed he calcula ion o he MD, as well as i s s an-
da d e o .
32
The NMA allows us o know he e ec size be ween any
pai o ea men s, whe he hey ha e been compa ed di ec ly o no . To
es ablish a anking be ween all ea men s, i is possible o assign a
p-sco e o each ea men . This sco e is based on he p obabili y ha a
ea men is be e han any o he ea men included in he NMA.
33–35
Planned me hods o analysis
The NMA de eloped in his s udy is amed wi hin he amewo k o
equen is s a is ics.
36,37
NMA is pe o med in RS udio using he ‘ne -
me a’ lib a y.
38
Cohe ence is he s a is ical ool o es he ansi i i y be ween
ea men e ec s.
33
The objec i e is o s udy whe he he e a e s a is-
ically signi ican di e ences be ween di ec and indi ec compa isons o
any wo ea men s. Fo his pu pose, he SIDE (Sepa a ing Indi ec om
Di ec E idence) me hod is used o s udy local cohe ence and he I
2
s a is ic is used in he case o global cohe ence.
33,39,40
This s a is ic is
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
2
calcula ed om Coch an’s Q.
40
The NMA is de eloped acco ding o a andom-e ec s model ha
conside s he he e ogenei y o he es ima es. Fu he mo e, he andom-
e ec s model acili a es ex apola ion o he esul s ob ained o a la ge
popula ion.
39,41
Resul s
Fig. 1 desc ibes he s udy selec ion p ocess. Six y a icles we e
included in he quali a i e syn hesis
42–101
and 51 we e included in he
quan i a i e syn hesis (Supplemen a y ma e ial -Table S.2).
Cha ac e is ics o he s udy and esul s o indi idual s udies
All s udies we e published be ween 1992 and 2021. Th ee s ud-
ies
102–104
had wo ea men a ms wi h he same exe cise app oach bu
di e en doses, so hey we e coun ed as six s udies in quan i a i e
syn hesis.
The sample size in each a icle was be ween 15 and 207, wi h an
a e age o 59 pa icipan s. The o al numbe o pa icipan s was 3581, o
which 2873 we e included in he quan i a i e syn hesis (708 no
included). The mean age was app oxima ely 48.9 ±6.1 yea s ( ange =
32–59 yea s). O he 60 s udies included, 48 used he ACR1990 c i e ia,
13 he ACR2010 c i e ia, and 2 he ACR2016 c i e ia. The o al numbe
o s udies and pa icipan s pe exe cise modali y, he o al numbe o
Fig. 1. Flow cha o he selec ion o s udies.
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
3
pa icipan s pe na ionali y and in e en ion me hod, and he diagnos ic
c i e ia used a e de ailed in Table S.4 (Supplemen a y ma e ial).
In e en ion g oup (The apeu ic exe cise)
The in e en ions we e classi ied acco ding o he de ini ions gi en
in Table S.5 (Supplemen a y ma e ial) and desc ibed acco ding o he
exe cise app oach used, he p og am design, and he dosage (in ensi y,
equency, and du a ion) (Supplemen a y ma e ial -Table S.2). . One o
he iden i ied ca ego ies ( ib a ion exe cise) was only in es iga ed in
one s udy.
105
Howe e , i did no p o ide he necessa y da a o be
included in he me a-analysis, so i was only included in he quali a i e
syn hesis. Thus, o he 15 TE ca ego ies iden i ied, 14 we e included in
he NMA.
Compa ison g oup (compa a o in e en ion)
Eigh in e en ion ca ego ies we e iden i ied as compa ison g oups
in he included s udies (Supplemen a y ma e ial -Table S.5). S udies ha
combined exe cise he apy wi h ano he ea men app oach we e
Fig. 2. Ne wo k plo p esen ing he ial da a con ibu ing e idence compa ing exe cise ea men ypes. A: Sho - e m (≤3 mon hs): 37 ials, 171 compa isons,
1922 pa icipan s. B: Long- e m (>3 mon hs): 24 ials, 136 compa isons, 1530 pa icipan s. The size o he nodes ep esen s how many imes he exe cise appea s in
any compa ison abou ha ea men and he wid h o he edges ep esen s he o al sample size in he compa isons i connec s.
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
4
classi ied as "Combined The apy" and he e o e quali ied as a compa i-
son g oup. S udies ha included a placebo ea men we e classi ied as
“sham” and hose in which no in e en ion o usual ca e was pe o med
as "usual ca e".
Ou comes
The main analyzes o his s udy e alua ed he e ec o he di e en
iden i ied TE app oaches on pain in ensi y in women wi h ib omyalgia,
assessed wi h VAS. All a ailable esul s we e collec ed o he a ailable
ollow-up ime poin s, wi h wo- ime cu -o poin s o he p ima y me a-
analyzes: sho - e m (≤3 mon hs) and long- e m (>3 mon hs).
Syn hesis o esul s
The ne wo k o med by he ea men s and hei compa isons is
cons uc ed om a g aph whe e each node ep esen s a ea men and
each edge a di ec compa ison.
106
The size o he nodes is p opo ional
o he sample collec ed o each ea men . Howe e , he hickness o
each edge is p opo ional o he numbe o compa isons collec ed be-
ween he wo ea men s a i s ends.
Pain esul s
Fo he sho - e m (Fig. 2A), a o al o 19 in e en ions we e
analyzed by 37 di e en s udies (35 s udies plus wo s udies ha had
wo ea men a ms o he same in e en ion). Among hem, he e we e
32 pai wise compa isons and i e compa isons be ween he h ee
in e en ion g oups a he same ime. Usual ca e had he la ges sample
size (n =307), ollowed by mixed exe cise (n =290), aqua ic exe cise (n
=236), lexibili y (n =234), and ae obic exe cise (n =233). The
hickness o he edges de e mines ha he e a e a g ea e numbe o
Fig. 3. T ea men ankings and su ace unde he cumula i e anking cu e (SUCRA) o pain in ensi y. A: Sho - e m (≤3 mon hs); B: Long- e m (>3 mon hs).
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
5
compa isons be ween aqua ic exe cise and mixed exe cise (4 compa i-
sons) and be ween lexibili y and esis ance aining (3 compa isons).
Rega ding ansi i i y, he SIDE me hod did no show s a is ically
signi ican di e ences be ween di ec and indi ec compa isons o
ea men s (Supplemen a y ma e ial -Table S.6A). Fu he mo e, he I
2
s a is ic had a alue o 84.7 % (78.5 %, 89.1 %).
Fig. 3 shows he anking be ween all ea men s. The p-sco e ep e-
sen ed o each TE in e en ion in Fig. 3A places he aqua ic exe cise
g oup in he i s posi ion (wi h a alue o 0.8713), ollowed by Pila es
(0.775).
Fig. 4 shows he ela i e e ec s be ween all pai s o compa isons and
hei espec i e con idence in e als a he 95 % le el. A nega i e alue
o he e ec size indica es he supe io i y o he column ea men , while
a posi i e alue indica es he supe io i y o he ow ea men . The o de
o he columns in Fig. 4A was cons uc ed ollowing he anking es ab-
lished in Fig. 3A.
S a is ically signi ican di e ences we e ound in a o o aqua ic
exe cise, Pila es, qigong, esis ance aining, i ual eali y, mixed ex-
e cise, and ae obic exe cise compa ed o usual ca e and sham. Aqua ic
exe cise also ob ained s a is ically signi ican di e ences in compa isons
wi h mixed exe cise, ae obic exe cise, lexibili y, and co e s abili y.
Taking he usual ca e as a e e ence, i is possible o obse e he e ec
size wi h espec o he o he in e en ions in Fig. 5A.
Finally, 12 sho - e m pai wise me a-analyses we e also pe o med
(Supplemen a y ma e ial -Figs. S1-S6 and S.9-S14), o which i e TE
app oaches showed s a is ically signi ican esul s in a o o aqua ic
exe cise (p <0.0001), Tai Chi (p <0.00001), i ual eali y (p =0.001),
esis ance aining (p =0.02), and qigong (p =0.02). On he o he hand,
an in e en ion ( lexibili y) had s a is ically signi ican di e ences in
a o o he compa ison g oup, wi h a p- alue <0.005 (p =0.002). All
esul s a e summa ized in Table S. 7 (Supplemen a y ma e ial).
Fo he long- e m (Fig. 2B), he e we e 17 in e en ions e alua ed by
24 s udies (22 s udies, and wo mo e wi h one pai o ea men a ms
each). O he 24 s udies, 20 made pai wise compa isons and ou p e-
sen ed compa isons be ween h ee ea men s a he same ime. Node
hickness again indica es ha he usual ca e g oup has he la ges
sample size (n =312), ollowed by he ae obic exe cise (n =177), mixed
exe cise (n =164), and aqua ic exe cise (n =156) g oups. Fu he mo e,
he numbe o compa isons be ween he usual ca e and ae obic exe cise
g oups was he highes , wi h a o al o ou be ween all s udies, which
esul ed in a highe hickness on he co esponding edge in Fig. 2B.
Rega ding ansi i i y, he ne spli ing me hod also did no show
s a is ically signi ican di e ences be ween di ec and indi ec com-
pa isons o ea men s (Supplemen a y ma e ial -Table S.6B). Fu he -
mo e, he I
2
s a is ic had a alue o 0 % (0 %, 56.6 %).
The p-sco e ep esen ed o each TE in e en ion in Fig. 3B places
esis ance aining i s (0.9749), ollowed by dance (0.905) and unc-
ional aining (0.8502).
The o de o he columns in Fig. 4B was also cons uc ed ollowing
he anking es ablished in Fig. 3B.Fig. 4B shows ha a g ea e numbe o
s a is ically signi ican compa isons we e ob ained in he long e m han
in he sho e m. Thus, we can highligh ha s a is ically signi ican
di e ences we e ound in a o o esis ance aining, dance, unc ional
aining, aqua ic exe cise, i ual eali y, and ae obic exe cise compa ed
o usual ca e, and s a is ically signi ican di e ences we e ound in a o
o esis ance aining, dance, unc ional aining, aqua ic exe cise, ae -
obic exe cise, and mixed exe cise compa ed o educa ion. Fu he mo e,
esis ance aining showed s a is ically signi ican di e ences in all
compa isons, excep compa ed o dance and unc ional aining. Fig. 5B
shows he e ec size using usual ca e as a e e ence g oup.
Finally, 10 pai wise me a-analyses we e pe o med in he long- e m
(Supplemen a y ma e ial -Figs. S.1-S.9 and S.13), o which six TE ap-
p oaches showed s a is ically signi ican esul s in hei a o : esis ance
aining (p <0.0001), aqua ic exe cise (p =0.0002), dance (p =0.003),
unc ional aining (p =0.004), ae obic exe cise (p =0.008), and i ual
eali y (p =0.04). As in he sho - e m, lexibili y had s a is ically
signi ican di e ences in a o o he compa ison g oup, wi h a p- alue
<0.005 (p =0.003). All esul s a e summa ized in Table S.7 (Supple-
men a y ma e ial).
Clinically impo an di e ences (CID)
Fig. 5 shows he CID o he in e en ions analyzed compa ed o usual
ca e. In he sho - e m (Fig. 4A), CIDs we e ob ained o many TE ap-
p oaches (aqua ic exe cise, Pila es, qigong, esis ance aining, i ual
eali y, mixed exe cise, Tai Chi, and ae obic exe cise). Howe e , in he
long e m (Fig. 4B), only h ee in e en ions achie ed a CID: esis ance
aining, dance, and unc ional aining. Fu he mo e, only esis ance
aining showed a mode a ely impo an clinical di e ence (≥30 poin s
ou o 100). Finally, in Fig. 4A and 4B, some compa isons be ween TE
app oaches and he compa a o s p esen ed CID in a o o he
compa a o s.
Risk o bias
Mos s udies (48 ou o 60) we e a ed as "good quali y" on he PED o
scale. Se en s udies we e a ed as "excellen quali y" and i e s udies
we e a ed as " ai quali y" (Supplemen a y ma e ial -Table S.8).
Publica ion bias isk assessmen
The Begg and Egge es s did no e eal s a is ical e idence o pub-
lica ion bias (p >0.05). These indings a e shown in he unnel plo s
(Supplemen a y ma e ial -Fig. S.15-S.28). The sensi i i y analysis indi-
ca ed ha he gene al esul s we e no subs an ially modi ied by elimi-
na ing any esul .
Ce ain y o he e idence
In he sho - e m, 171 compa isons (47 s udies) we e pe o med, o
which 22 we e a ed as “mode a e quali y”. The emaining 149 com-
pa isons we e a ed as "low quali y". Long- e m, 137 compa isons (32
s udies) we e pe o med, o which only one was a ed as "high quali y"
(ae obic exe cise e sus usual ca e). O he emaining 136 compa isons,
110 we e a ed as "mode a e quali y" and 26 as "low quali y". (Supple-
men a y ma e ial -Table S.98).
Discussion
To ou knowledge, his is he i s NMA o compa e di e en TE
app oaches used in women wi h ib omyalgia o educe pain in ensi y.
Ou NMA es ablished a anking (Fig. 3) o TE app oaches ha we used o
o de he columns in Fig. 4 showing he ela i e e ec s be ween all pai s
o compa isons. The e o e, Fig. 4 can be used as a aluable decision-
making ool when choosing he mos app op ia e ype o TE o educe
pain in ensi y in women wi h ib omyalgia.
In he sho - e m (Fig. 4A), we had se en TE app oaches ha
gene a ed s a is ically signi ican di e ences compa ed o usual ca e
and sham; also, hese imp o emen s we e clinically impo an . I we
analyze he compa isons be ween hese se en app oaches, ou esul s
appea o indica e ha , in he sho e m, he i e mos e ec i e ype o
TE o imp o e pain in ensi y in women wi h ib omyalgia we e aqua ic
exe cise, Pila es, qigong, esis ance aining, and i ual eali y.
Fu he mo e, he s eng h o e idence, in all hese compa isons wi h
usual ca e was mode a e, excep o Pila es, which was low.
In he long- e m (Fig. 4B), we had six TE app oaches ha gene a ed
s a is ically signi ican di e ences compa ed o usual ca e and educa-
ion. Compa ed o sham, i e ypes o TE gene a ed s a is ically signi -
ican di e ences. I we analyze he compa isons be ween hese exe cise
app oaches, ou esul s appea o indica e ha , in he long- e m, he
h ee mos e ec i e ype o TE, compa ed o usual ca e, educa ion, and
sham, o imp o e pain in ensi y in women wi h ib omyalgia, we e
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
6
Fig. 4. League able o ne wo k me a-analysis esul s o all compa isons be ween exe cise and non-exe cise in e en ions. E ec s a e exp essed as he mean di -
e ence (95 % CI) be ween in e en ions. A: sho - e m ou comes (≤3 mon hs); B: long- e m esul s (>3 mon hs).
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
7
esis ance aining, dance, and unc ional aining. Fu he mo e, esis-
ance aining and dance we e he mos e ec i e TE compa ed o d ug
he apy, acupunc u e, and combina ion he apy; and only esis ance
aining was he mos e ec i e TE compa ed o elaxa ion. All o hese
imp o emen s we e clinically impo an . The s eng h o e idence, o all
hese compa isons was mode a e.
Impo an ly, esis ance aining was he only in e en ion ha
showed s a is ically signi ican and clinically ele an di e ences
compa ed o usual ca e and sham in bo h he sho and long e m. I was
also he only in e en ion ha showed a mode a ely impo an clinical
di e ence (≥30 poin s ou o 100), so, o e all, esis ance aining may
be he mos e ec i e ype o TE o educe pain in ensi y in women wi h
ib omyalgia.
Ou indings ha e signi ican clinical implica ions o se e al ea-
sons. Mos impo an ly, Fig. 4 can se e as a aluable guide o selec ing
he mos app op ia e TE app oaches o educe pain in ensi y in women
wi h ib omyalgia. This decision should also be made based on he pa-
ien ’s cha ac e is ics and p e e ences, as his is a key ac o in
imp o ing adhe ence o ea men .
107–109
These indings could also
con ibu e o a change in he cu en pa adigm ega ding TE p esc ip-
ion in ib omyalgia, because he "gold s anda d" ecommenda ion o
ae obic exe cise should be eplaced by o he mo e e ec i e app oaches.
Speci ically, aqua ic exe cise was mo e e ec i e han ae obic exe cise in
he sho - e m and esis ance aining and dance we e mo e e ec i e in
he long- e m.
This e iew has se e al limi a ions. Fi s , s udies a e a isk o bias
due o he impossibili y o blinding when ac i e in e en ions a e used.
Second, his e iew is limi ed o he e ec s o exe cise in women wi h
ib omyalgia. I seems mo e app op ia e o limi he sample o his sex,
as ib omyalgia a ec s mainly women and including men in he e iew
would inc ease he isk o bias. The e o e, hese esul s canno be
ex apola ed o men wi h ib omyalgia. Thi d, he e a e conce ns abou
he pa ien selec ion c i e ia used in RCTs. A o al o 80 % o he ials
used he ACR 1990 c i e ia, he alidi y o which was e u ed by he
au ho s hemsel es in 2010. Despi e his, s udies published in 2021 s ill
con inue o use hese c i e ia.
110–112
This could lead o se ious selec ion
bias in pa ien s, which could a ec he ex apola ion o he esul s o he
clinical se ing. Finally, ou s udy has e alua ed he e icacy o di e en
TE app oaches in educing pain in ensi y. Howe e , i should be no ed
ha in ensi y is only one dimension o pain expe ience, and he e o e an
imp o emen in pain in ensi y does no necessa ily imply an imp o e-
men in he o e all impac o he disease.
19
Fu u e s udies should ocus on whe he di e en pa ame e s o he
FITT model ( equency, in ensi y, ime, and ype), could op imize ou -
comes in educing pain in ensi y in women wi h ib omyalgia. Finally, i
is impe a i e o uni y diagnos ic c i e ia in ib omyalgia so ha hese
pa ien s a e classi ied mo e accu a ely and homogeneously. The c i e ia
cu en ly used o he diagnosis o ib omyalgia a e s ill based on a
biomedical model, igno ing he ecommenda ions o he Wo ld Heal h
O ganisa ion (WHO). Psychosocial a iables a e known o ha e a sig-
ni ican impac on disease de elopmen , so i is necessa y o es ablish
diagnos ic c i e ia based on a biopsychosocial model.
113
Conclusions
The NMA showed, wi h a mode a e le el o e idence, ha he mos
e ec i e TE app oach o educe pain in ensi y in women wi h ib o-
myalgia was, in he sho - e m, aqua ic exe cise and, in he long- e m,
esis ance aining. Fu he mo e, esis ance aining was he only
in e en ion ha showed sho - and long- e m imp o emen s, wi h a
mode a ely impo an clinical di e ence. Mo e RCTs a e needed o
s eng hen hese indings.
Decla a ion o compe ing in e es
The au ho s decla e no compe ing in e es .
Acknowledgemen s
We app ecia e he wo k o all he au ho s who conduc ed he clinical
ials ha we e included in ou me a-analysis.
Supplemen a y ma e ials
Supplemen a y ma e ial associa ed wi h his a icle can be ound, in
he online e sion, a doi:10.1016/j.bjp .2025.101226.
Re e ences
1. And ade A, Dominski FH, Sieczkowska SM. Wha we al eady know abou he
e ec s o exe cise in pa ien s wi h ib omyalgia: an umb ella e iew. Semin
Fig. 5. Summa y ne wo k me a-analysis esul s o each exe cise ype compa ed wi h Usual Ca e. A: sho - e m (≤3 mon hs); B: long e m (>3 mon hs). The hashed
line indica es clinically impo an di e ence.
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
8
A h i is Rheum. 2020;50(6):1465–1480. h ps://doi.o g/10.1016/J.
SEMARTHRIT.2020.02.003.
2. Mac a lane GJ, K onisch C, Dean LE, e al. EULAR e ised ecommenda ions o
he managemen o ib omyalgia. Ann Rheum Dis. 2017;76(2):318–328. h ps://
doi.o g/10.1136/ANNRHEUMDIS-2016-209724.
3. Sa zi-Pu ini P, Gio gi V, Ma o o D, A zeni F. Fib omyalgia: an upda e on clinical
cha ac e is ics, ae iopa hogenesis and ea men . Na Re Rheuma ol. 2020;16(11):
645–660. h ps://doi.o g/10.1038/s41584-020-00506-w, 2020 16:11.
4. Ma ques AP, San o A de S do E, Be ssane i AA, Ma su ani LA, Yuan SLK. P e alence
o ib omyalgia: li e a u e e iew upda e. Re B as Reuma ol. 2017;57(4):356–363.
h ps://doi.o g/10.1016/J.RBRE.2017.01.005.
5. Cabo Mesegue A, Ce d´
a-Olmedo G, T illo Ma a JL. Fib omyalgia: p e alence,
epidemiologic p o iles and economic cos s. Med Clin. 2017;149(10):441–448.
6. Gal ez-S´
anchez CM, Duschek S, Del Paso GAR. Psychological impac o
ib omyalgia: cu en pe spec i es. Psychol Res Beha Manag. 2019;12:117.
h ps://doi.o g/10.2147/PRBM.S178240.
7. Wol e F, Wali B, Pe o S, Raske JJ, H¨
ause W. Fib omyalgia diagnosis and
biased assessmen : sex, p e alence and bias. PLoS One. 2018;13(9), e0203755.
8. Wali B, Ka z RS, Be gman MJ, Wol e F. Th ee-qua e s o pe sons in he US
popula ion epo ing a clinical diagnosis o ib omyalgia do no sa is y
ib omyalgia c i e ia: he 2012 Na ional Heal h In e iew Su ey. PLoS One. 2016;
11(6), e0157235.
9. Sil e wood V, Chew-G aham CA, Raybould I, Thomas B, Pe e s S. I i ’s a medical
issue i would ha e co e ed i by now”: lea ning abou ib omyalgia h ough he
hidden cu iculum: a quali a i e s udy. BMC Med Educ. 2017;17(1):1–9. h ps://
doi.o g/10.1186/S12909-017-0972-6/TABLES/1.
10. Yunus MB. Cen al Sensi i i y synd omes: a new pa adigm and g oup nosology o
ib omyalgia and o e lapping condi ions, and he ela ed issue o disease e sus
illness. Semin A h i is Rheum. 2008;37(6):339–352. h ps://doi.o g/10.1016/J.
SEMARTHRIT.2007.09.003.
11. Cagnie B, Coppie e s I, Denecke S, Six J, Danneels L, Meeus M. Cen al
sensi iza ion in ib omyalgia? A sys ema ic e iew on s uc u al and unc ional
b ain MRI. Semin A h i is Rheum. 2014;44(1):68–75. h ps://doi.o g/10.1016/J.
SEMARTHRIT.2014.01.001.
12. Meeus M, Nijs J. Cen al sensi iza ion: a biopsychosocial explana ion o ch onic
widesp ead pain in pa ien s wi h ib omyalgia and ch onic a igue synd ome. Clin
Rheuma ol. 2006;26(4):465–473.
13. Bidonde J, Busch AJ, Webbe SC, e al. Aqua ic exe cise aining o ib omyalgia.
Coch ane Da abase Sys Re . 2014;(10). h ps://doi.o g/10.1002/14651858.
CD011336/MEDIA/CDSR/CD011336/IMAGE_N/NCD011336-CMP-012-02.PNG,
2014.
14. Pa k JH, Nie mann KJ, Olsen N. E idence o me abolic abno mali ies in he
muscles o pa ien s wi h ib omyalgia. Cu Rheuma ol Rep 2000 2:2. 2000;2(2):
131–140.
15. H¨
ause W, Klose P, Langho s J, e al. E icacy o di e en ypes o ae obic exe cise
in ib omyalgia synd ome: a sys ema ic e iew and me a-analysis o andomised
con olled ials. A h i is Res The . 2010;12(3):1–14. h ps://doi.o g/10.1186/
AR3002/TABLES/5.
16. Busch AJ, Webbe SC, Richa ds RS, e al. Resis ance exe cise aining o
ib omyalgia. Coch ane Da abase Sys Re . 2013;(12). h ps://doi.o g/10.1002/
14651858.CD010884/MEDIA/CDSR/CD010884/IMAGE_N/NCD010884-CMP-
006-06.PNG, 2013.
17. Nelson NL. Muscle s eng hening ac i i ies and ib omyalgia: a e iew o pain and
s eng h ou comes. J Bodyw Mo The . 2015;19(2):370–376. h ps://doi.o g/
10.1016/J.JBMT.2014.08.007.
18. Bidonde J, Busch AJ, Schach e CL, e al. Ae obic exe cise aining o adul s wi h
ib omyalgia. Coch ane Da abase Sys em Re . 2017;2017(6). h ps://doi.o g/
10.1002/14651858.CD012700/MEDIA/CDSR/CD012700/IMAGE_N/NCD012700-
CMP-004-06.PNG.
19. Sosa-Reina MD, Nunez-Nagy S, Gallego-Izquie do T, Pecos-Ma ín D, Monse a J,
´
Al a ez-Mon M. E ec i eness o he apeu ic exe cise in Fib omyalgia synd ome: a
sys ema ic e iew and me a-analysis o andomized clinical ials. Biomed Res In .
2017. h ps://doi.o g/10.1155/2017/2356346, 2017.
20. And ade A, de Aze edo Klumb S e ens R, Sieczkowska SM, Pey ´
e Ta a uga LA,
To es Vila ino G. A sys ema ic e iew o he e ec s o s eng h aining in pa ien s
wi h ib omyalgia: clinical ou comes and design conside a ions. Ad Rheuma ol.
2019;58(1):36. h ps://doi.o g/10.1186/S42358-018-0033-9.
21. Bidonde J, Busch AJ, Schach e CL, e al. Mixed exe cise aining o adul s wi h
ib omyalgia. Coch ane Da abase Sys Re . 2019;2019(5). h ps://doi.o g/10.1002/
14651858.CD013340/MEDIA/CDSR/CD013340/IMAGE_N/NCD013340-CMP-
004-06.PNG.
22. B a o C, Skjae en LH, Gui a d Sein-Echaluce L, Ca alan-Ma amo os D.
E ec i eness o mo emen and body awa eness he apies in pa ien s wi h
ib omyalgia: a sys ema ic e iew and me a-analysis. Eu J Phys Rehabil Med. 2019;
55(5):646–657. h ps://doi.o g/10.23736/S1973-9087.19.05291-2.
23. Busch AJ, Ba be KAR, O e end TJ, Peloso PMJ, Schach e CL. Exe cise o ea ing
ib omyalgia synd ome. Coch ane Da abase Sys Re . 2007. h ps://doi.o g/
10.1002/14651858.CD003786.PUB2/MEDIA/CDSR/CD003786/IMAGE_N/
NCD003786-CMP-002-05.PNG. Published online.
24. Hu on B, Salan i G, Caldwell DM, e al. The PRISMA Ex ension S a emen o
epo ing o sys ema ic e iews inco po a ing ne wo k me a-analyses o Heal h
ca e in e en ions: checklis and explana ions. Ann In e n Med. 2015;162(11):
777–784. h ps://doi.o g/10.7326/M14-2385.
25. Hozo SP, Djulbego ic B, Hozo I. Es ima ing he mean and a iance om he
median, ange, and he size o a sample. BMC Med Res Me hodol. 2005;5(1):1–10.
h ps://doi.o g/10.1186/1471-2288-5-13/TABLES/3.
26. Wan X, Wang W, Liu J, Tong T. Es ima ing he sample mean and s anda d
de ia ion om he sample size, median, ange and/o in e qua ile ange. BMC
Med Res Me hodol. 2014;14(1):1–13. h ps://doi.o g/10.1186/1471-2288-14-135/
TABLES/3.
27. Mahe CG, She ing on C, He be RD, Moseley AM, Elkins M. Reliabili y o he
PED o scale o a ing quali y o andomized con olled ials. Phys The . 2003;83
(8):713–721. h ps://doi.o g/10.1093/PTJ/83.8.713.
28. Ghai S, Ghai I, Schmi z G, E enbe g AO. E ec o hy hmic audi o y cueing on
pa kinsonian gai : a sys ema ic e iew and me a-analysis. Sci Rep. 2018;8(1):1–19.
h ps://doi.o g/10.1038/s41598-017-16232-5.
29. Papakons an inou T, Nikolakopoulou A, Higgins JPT, Egge M, Salan i G. CINeMA:
so wa e o semiau oma ed assessmen o he con idence in he esul s o ne wo k
me a-analysis. Campbell Sys Re . 2020;16(1):e1080. h ps://doi.o g/10.1002/
CL2.1080.
30. Dwo kin R, Tu k D, Wy wich K, e al. In e p e ing he clinical impo ance o
ea men ou comes in ch onic pain clinical ials: IMMPACT ecommenda ions.
J Pain. 2008;9(2):105–121.
31. Alb igh J, Allman R, Bon iglio R, e al. Philadelphia Panel e idence-based clinical
p ac ice Guidelines on selec ed ehabili a ion in e en ions: o e iew and
me hodology. Phys The . 2001;81(10):1629–1640.
32. Deeks J, Higgins J. S a is ical algo i hms in e iew manage 5. S a Algo i hm Re
Manage 5. 2007. Published online.
33. Ha e M, Cuijpe s P, Fu ukawa TA, Ebe DD. Doing me a-analysis wi h R : a
hands-on guide. Doing Me a-Anal R. 2021. h ps://doi.o g/10.1201/
9781003107347. Published online.
34. Salan i G, Ades AE, Ioannidis JPA. G aphical me hods and nume ical summa ies
o p esen ing esul s om mul iple- ea men me a-analysis: an o e iew and
u o ial. J Clin Epidemiol. 2011;64(2):163–171. h ps://doi.o g/10.1016/J.
JCLINEPI.2010.03.016.
35. Rücke G, Schwa ze G. Ranking ea men s in equen is ne wo k me a-analysis
wo ks wi hou esampling me hods. BMC Med Res Me hodol. 2015;15(1):58.
h ps://doi.o g/10.1186/s12874-015-0060-8.
36. Shim S, Lee J. Ne wo k me a-analysis: applica ion and p ac ice using R so wa e.
Epidemiol Heal h. 2019:41.
37. Hong H, Ca lin BP, Shamliyan TA, e al. Compa ing bayesian and equen is
app oaches o mul iple ou come mixed ea men compa isons. Med Decis Making.
2013;33(5):702–714. h ps://doi.o g/10.1177/0272989×13481110.
38. Rücke G, K ahn U, K¨
onig J, E himiou O, Schwa ze G. Ne me a: ne wo k me a-
analysis using equen is me hods. R Package Ve sion. 2019;1(2).
39. Viech baue W. Accoun ing o he e ogenei y ia andom-e ec s models and
mode a o analyses in me a-analysis. J Psychol. 2007;215(2):104–121. h ps://doi.
o g/10.1027/0044-3409.215.2.104.
40. Higgins J, Thompson S, Deeks J, Al man D. Measu ing inconsis ency in me a-
analyses. BMJ. 2003;327(7414):557–560. h ps://doi.o g/10.1136/
bmj.327.7414.557.
41. Te Vee E, an Oijen MGH, an Laa ho en HWM. The use o (Ne wo k) me a-
analysis in clinical oncology. F on Oncol. 2019;9(AUG):822. h ps://doi.o g/
10.3389/FONC.2019.00822/BIBTEX.
42. Mengshoel AM, Komnaes HB, Fø e O. The e ec s o 20 weeks o physical i ness
aining in emale pa ien s wi h ib omyalgia. Clin Exp Rheuma ol. 1992;10(4):
345–349. h p://www.ncbi.nlm.nih.go /pubmed/1395219.
43. H¨
akkinen A, H¨
akkinen K, Hannonen P, Alen M. S eng h aining induced
adap a ions in neu omuscula unc ion o p emenopausal women wi h
ib omyalgia: compa ison wi h heal hy women. Ann Rheum Dis. 2001;60(1):21–26.
h ps://doi.o g/10.1136/a d.60.1.21.
44. Jen o ES, K al ik AG, Mengshoel AM. E ec s o pool-based and land-based
ae obic exe cise on women wi h ib omyalgia/ch onic widesp ead muscle pain.
A h i is Rheum. 2001;45(1):42–47. h ps://doi.o g/10.1002/1529-0131(200102)
45:1<42. :AID−ANR82>3.0.CO;2-A.
45. Jones KD, Bu ckha d CS, Cla k SR, Benne RM, Po empa KM. A andomized
con olled ial o muscle s eng hening e sus lexibili y aining in ib omyalgia.
J Rheuma ol. 2002;29(5):1041–1048. h p://www.ncbi.nlm.nih.go /pubmed
/12022321.
46. an San en M, Bolwijn P, Landew´
e R, e al. High o low in ensi y ae obic i ness
aining in ib omyalgia: does i ma e ? J Rheuma ol. 2002;29(3):582–587. h p
://www.ncbi.nlm.nih.go /pubmed/11908577.
47. Schach e CL, Busch AJ, Peloso PM, Sheppa d MS. E ec s o sho e sus long
bou s o ae obic exe cise in seden a y women wi h ib omyalgia: a andomized
con olled ial. Phys The . 2003;83(4):340–358. h p://www.ncbi.nlm.nih.go /p
ubmed/12665405.
48. Valim V, Oli ei a L, Suda A, e al. Ae obic i ness e ec s in ib omyalgia.
J Rheuma ol. 2003;30(5):1060–1069. h p://www.ncbi.nlm.nih.go /pubmed
/12734907.
49. Sencan S, Ak S, Ka an A, Muslumanoglu L, Ozcan E, Be ke E. A s udy o compa e
he he apeu ic e icacy o ae obic exe cise and pa oxe ine in ib omyalgia
synd ome. J Back Musculoskele Rehabil. 2004;17(2):57–61. h ps://doi.o g/
10.3233/BMR-2004-17204.
50. Valkeinen H. Changes in knee ex ension and lexion o ce, EMG and unc ional
capaci y du ing s eng h aining in olde emales wi h ib omyalgia and heal hy
con ols. Rheuma ology (Ox o d). 2003;43(2):225–228. h ps://doi.o g/10.1093/
heuma ology/keh027.
51. Gusi N, Tomas-Ca us P, H¨
akkinen A, H¨
akkinen K, O ega-Alonso A. Exe cise in
wais -high wa m wa e dec eases pain and imp o es heal h- ela ed quali y o li e
and s eng h in he lowe ex emi ies in women wi h ib omyalgia. A h i is Rheum.
2006;55(1):66–73. h ps://doi.o g/10.1002/a .21718.
´
A.-J. Rod íguez-Domínguez e al.
B azilian Jou nal o Physical The apy 29 (2025) 101226
9
