efficacy of hypertonic dextrose infiltrations for pain
TRANSCRIPT
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
156
ORiGinAL ARTiCLESREViEw ARTiCLES
1. División Académica Multidisciplinaria de Comalcalco, Comalcalco,
Tabasco, México; Universidad Juárez Autónoma de Tabasco
2. Centro Médico Olympia, Cancún, México
3. Universidad Juárez Autónoma de Tabasco, División Académica
de Ciencias de la Salud, Villahermosa, Tabasco, México; División
Académica Multidisciplinaria de Jalpa de Méndez, Jalpa de
Méndez, Tabasco, México
4. Hospital General de Yajalón, Secretaría de Salud. Yajalón,
Chiapas, México
5. División Académica de Ciencias de la Salud, Villahermosa,
Tabasco, México; Universidad Juárez Autónoma de Tabasco;
6. Universidad Autónoma de Yucatán, Mérida Yucatán, México
rious adverse events were observed when HDI wereused. Conclusions. We found that HDI reduced long termpain in individuals with rotator cuff. HDI could be analternative to non-invasive treatments when no favor-able results can be achieved. However, due to the smallnumber of studies included in this meta-analysis, newstudies are necessary to clarify the efficacy and safety ofthis intervention.
Keywords: Hypertonic dextrose; Prolotherapy; Infil-trations; Shoulder; Cuff rotator.
INTRODUCTION
The pathology of the rotator cuff has been consideredas the main cause of pain and disability of the shoulder1.The prevalence of injuries of the rotator cuff tendonsranges from 6% to 30%, increasing progressively withage1. For the treatment of this pathology conservativemodalities are commonly used including anti-inflam-matory drugs2, physical therapeutic modalities (3), ex-ercise programs (4), intra-articular and subacromial in-filtrations5, 6 and surgical procedures7, 8.
Regarding the different infiltration treatments, themost widely used is the application of corticosteroids5.Other infiltrations include the use of plasma rich inplatelets (PRP)5,6,9, hyaluronic acid5, 10, hypertonic dex-trose5, 6, botulinum toxin5, mesenchymal cells of bonemarrow6, 11 and a mixture of oxygen - medicinal ozone12.
On the other hand, injection therapy with sclerosingagents or irritant substances has been used for decadesas a complementary treatment for chronic muscu-loskeletal conditions. Dr. George Hackett defined theterm prolotherapy in the 1950s13, 14; this term involvesinjections of a solution with sclerosing agents or irri-
Efficacy of hypertonic dextrose infiltrations for pain control in rotator cuff tendinopathy:
systematic review and meta-analysis
Arias-Vázquez PI1, Tovilla-Zárate CA1, González-Graniel K1, Burad-Fonz W2, González-Castro TB3, López-Narváez ML4, Castillo-Avila RG5, Arcila-Novelo R6
ACTA REUMATOL PORT. 2021;46:156-170
ABSTRACT
Introduction.Our study aimed to assess the efficacy ofhypertonic dextrose infiltrations (HDI) for pain controlin individuals with rotator cuff tendinopathy and to as-sess the characteristics of the treatment and the pre -sence of side effects or adverse reactions through a sys-tematic review and meta-analysis. Methods. The search for the articles was performed inthe electronic databases PUBMED, EMBASE, SCOPUS,SCIELO, DIALNET and Google Scholar, published up toAugust 2020. The keywords used were "prolotherapy" or“proliferation therapy” or "hypertonic dextrose infiltra-tions" or "hypertonic dextrose injection" and "RotatorCuff" or "Rotator Cuff Injury" or “Rotator Cuff Tear” or“Rotator Cuff Tendinosis” or "supraspinatus". The effec-tiveness of HDI was expressed as standardized mean dif-ference (d) and 95% CI. Results. In the pooled analysis, HDI were an effectiveintervention to reduce long-term pain in patients withrotator cuff tendinopathy when compared to controls;furthermore, in the individual analyses, HDI were moreeffective in the short, medium and long terms than non-invasive treatments, and more effective in the long-termthan infiltrations with local anesthetics. On the otherhand, HDI were not more effective than injections withcorticosteroids or PRP. Finally, no complications or se -
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
157
AriAs-VAzquez Pi et Al
tating properties in the ligament-bone or tendon-boneareas or the intra-articular space, performed repeated-ly at established intervals13, 14. The most common pro-lotherapy agent used in the clinical practice is the hy-pertonic dextrose solution at concentrations rangingfrom 12.5% to 25%, applied as intra-articular and/orextra-articular infiltrations on ligament and tendon in-sertions, to favor the repair processes of the affectedtissues13, 14. Hypertonic dextrose prolotherapy has beenreported to be effective for treating knee osteoarthritiswhere it has been reported to be more effective than in-filtrations with local anesthetics, as effective as infil-trations with hyaluronic acid, ozone or radiofrequen-cy, and less effective than PRP without side effects15.When used for treating tendinopathies of the lowerlimb it has been reported to be a safe and effective treat-ment for Achilles tendinopathy, plantar fasciitis andOsgood-Schlatter disease16, 17. When hypertonic dex-trose has been used for treating upper limb pathologiessuch as hand osteoarthritis, lateral epicondylitis androtator cuff disease, clinical studies have reported pos-itive results without side effects17, 18. However, HDI re-mains a controversial therapy for treating rotator cufftendinopathy and it is classified as a complementarytherapy. Therefore, the objective of our study was toperform a systematic search of clinical studies that usedHDI in patients with rotator cuff pathology, to analyzeits efficacy in pain control, the characteristics of thetreatment and the presence of side effects or adverse re-actions through a meta-analysis.
MeTHODOLOGy
The methodology used was based on the Preferred Re-porting Items for Systematic Reviews and Meta-Ana -lyses (PRISMA) guidelines19 for the presentation of sys-tematic review and meta-analysis.
MeTHODS AND SeARCH STRATeGy
Articles of interest were identified in electronicdatabases using a search period up to August 2020.The databases used were PUBMED, EMBASE, SCO-PUS, SCIELO, DIALNET and gray literature as GoogleScholar. The search terminology included the MESHterms (and entry terms) "prolotherapy" or “prolifera-tion therapy” or "hypertonic dextrose infiltrations" or"hypertonic dextrose injection" and "rotator cuff" or"rotator cuff injury" or “rotator cuff tear” or “rotatorcuff tendinosis” or "supraspinatus", as well as multiple
combinations between these terms. The search of eligible studies was performed with-
out language restrictions. After searching in thedatabases above mentioned, a hand search of the ref-erence list in the articles and reviews was conducted tofind additional eligible studies.
TypeS Of STUDIeS
This review included randomized controlled trials(RCTs) and observational studies (cases-controls, se-ries of cases) that used HDI as a therapeutic interven-tion for the treatment of pain in individuals with rota-tor cuff tendinopathy. In the RCTs HDI were comparedagainst physiotherapy, exercise programs or againstother infiltrations (placebo or other therapeutic sub-stances). We excluded reviews, one-case reports, stud-ies of shoulder pathologies other than rotator cufftendinopathy or unspecified pathologies. The studiesselected had to describe in detail the interventions car-ried out, forms of evaluation and their results.
pARTICIpANTS
The selected studies included patients with rotator cufftendinopathy and met the following criteria:• Adults of at least 18 years of age.• Clinical and imaging (ultrasonography or magnetic re -
sonance imaging) diagnosis of rotator cuff tendinopa-thy (tendinosis, partial tear or full-thickness tear).
• Presence of pain and functional alterations of morethan 3 months of evolution.
• Participants treated with HDI and compared withother interventions.
Type Of INTeRveNTIONS
The selected studies included patients with rotator cufftendinopathy who were treated with HDI and com-pared with other interventions.
The criteria for the type of interventions used in thepatients of study groups were:• One or more treatment sessions with HDI (at a con-
centration greater than 10%).• The infiltrations applied in the rotator cuff tendons
insertion and/or intratendinous application in thefocal area in case of rupture and/or in subacromialor intra-articular space.
• The infiltrations performed following the anatomi-cal technique or under ultrasound guidance.
• Patients in the control groups were treated withphysiotherapy, exercise programs or infiltrations ofother substances.
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
158
HyPertonic dextrose infiltrAtion in rotAtor cuff tendinoPAtHy
• Co-interventions were allowed as long as they wereuniform in all groups.
evALUATION Of THe RISk Of BIAS AND THe
MeTHODOLOGICAL qUALITy Of THe INCLUDeD
STUDIeS
Two researchers independently assessed the metho -dological quality and risk of bias of each included study.The evaluation of the clinical trials was based on theCochrane Handbook for Systematic Reviews recom-mendations, version 5.120. The assessment of the riskof bias in non-randomized observational studies wasperformed using the ROBINS-I tool21. The rating of thele vel of evidence for therapeutic studies was deter-mined for each study using the scale of the AmericanSociety of Surgeons22.
evALUATION Of eLIGIBILITy AND DATA exTRACTION
Two researchers independently examined titles, abs -tracts and full texts, then determined the eligibility ofeach study. Disagreements were solved by consensusthrough the opinion of a third researcher. For eligiblestudies, data were extracted independently and in-cluded: study design, risk of bias, clinical configura-tion, characteristics of the participants, characteristicsof the interventions, results, duration of follow-up andadverse events.
The efficacy of HDI in pain control was establishedas the primary endpoint. Pain control was measuredby the visual analog scale (VAS) and was inclu ded in thequantitative analysis.
The evaluation of the improvement in function, thecharacteristics of the treatment and the adverse effects wereestablished as secondary endpoints and were des cribedin the qualitative analysis according to the data providedin the included studies. Improvement in function wasmeasured in terms of validated function scales such asShoulder Pain and Disability Index (SPADI), AmericanShoulder and Elbow Surgeons Standardized (ASES) andDisability of Arm and Shoulder Score (DASH).
The follow-up time was evaluated in the short (≤ 6weeks), medium (12 weeks) and long terms (≤ 24weeks).
STATISTICAL ANALySIS
For the evaluation of RCTs, studies were grouped ac-cording to the follow-up time. The effectiveness of HDIin control pain were expressed as standardized meandifference (d) and 95% CI. The standardized mean dif-ference was calculated comparing the study group ver-
sus (vs.) the comparison group. Heterogeneity acrossstudies was measured using Q statistic and inconsis-tence index I2. When p (Q) was <0.10, the presence ofheterogeneity was considered. When I2 > 50% largeheterogeneity was determined; when I2 = 25-50%moderate heteroge neity was considered, and when I2< 25% absence of heterogeneity was determined. Thepublication bias was evaluated by Begg’s funnel plotsgraphically and Egger’s test quantitatively. For Begg’sfunnel plots an asymmetry was considered as a signif-icant presence of bias. For the Egger's test, the signifi-cance was fixed as p <0.05. The meta-analysis was per-formed using EPIDAT 3.1 Software.
To evaluate the characteristics of the treatment andadverse effects, they were summarized in descriptivemeasures, according to the data provided in the in-cluded studies.
ReSULTS
A total of 116 citations were identified and 63 dupli-cates were excluded. Titles and abstracts of the re-maining 53 studies were read; then, 24 studies thatcontained animal models, other tendinopathies, edito-rials, comments and others were also excluded. Of the29 remaining studies, 21 were additionally excludedfor the following reasons: review studies (n = 16), stu -dies of shoulder pathologies other than rotator cufftendinopathy (n = 3), as well as one-case reports (n = 2). Finally, six clinical trials (23-28) and two obser-vational studies29,30 were eligible for inclusion in thissystematic review. The flowchart of the systematizedsearch is shown in Figure 1.
The final six RCTs selected included 157 indivi dualswith rotator cuff tendinopathy treated with HDI, per-formed in the tendon insertion area or in the focal areaof rupture of the rotator cuff tendons and/or in sub-acromial or intra-articular space, while 236 controlswere treated with exercise programs or infiltrationswith corticosteroids, lidocaine, platelet-rich plasma orsaline solution. Regarding the observational studies, 78individuals with rotator cuff tendinopathy were trea tedwith HDI performed with the same characteristics, using 53 controls.
Of the RCTs included in the qualitative analysis, twostudies showed a low risk of bias23, 26, three had amodera te risk of bias24,25,27 and one of them showed ahigh risk of bias28. The two observational studies in-cluded in the qualitative analysis showed a moderate
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
159
AriAs-VAzquez Pi et Al
fIGURe 1. Flow diagram of the systematic review
risk of bias29,30. Only RCTs with low or moderate riskof bias were included in our quantitative analysis. Thedesign characteristics and the risk of bias assessment ofthe included studies are summarized in Table I, Figure2 and Figure 3.
In the five RCTs included in the quantitative analy-sis23-27, all the groups studied were treated uniformlywith an exercise program as a co-intervention. Simi-larly, in these studies the use of NSAIDs during treat-ment or follow-up was restricted; the use of analgesicssuch as acetaminophen or tramadol was allowed in caseof post-infiltration pain.
The characteristics of the intervention and results ofeach study included are reported in Table II and III.
MeTA-ANALySIS Of THe effICACy Of
INfILTRATIONS wITH HypeRTONIC DexTROSe
fOR pAIN CONTROL IN ROTATOR CUff pATHOLOGy
In the five clinical trials included in this meta-analy-sis23-27, the treatment with HDI was compared with other interventions such as infiltration with local anes-thetics23,27, exercise programs24, infiltration with corti-costeroids25, 27, infiltration with saline solution26 and infiltrations with platelet-rich plasma27. The meta-anal-ysis was performed by time of follow-up and expressedit in terms of standardized mean difference.
Short–term follow-up: In the pooled analysis, no sta-tistically significant difference in pain reduction wasfound when comparing HDI vs. controls (d = -0.045,
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
160
HyPertonic dextrose infiltrAtion in rotAtor cuff tendinoPAtHy
TA
BLe
I. C
HA
RA
CT
eR
IST
ICS
Of
TH
e S
TU
Dy
De
SIG
N
Lev
el
Au
thor
Ye
arSa
mp
le s
ize
Stu
dy
Des
ign
All
ocat
ion
Bli
nd
ing
Evi
den
ceBer
tran
d et
al.
(23)
2016
47 p
atie
nts
wit
h c
hro
nic
ten
dinop
athy
of
Exp
erim
enta
l,R
ando
miz
ed t
o 3
Blindi
ng
of p
arti
cipa
nts
,I
rota
tor
cuff (
tendi
nos
is, p
arti
al t
ear
or
Pro
spec
tive
, tr
eatm
ent
grou
ps.
pers
onnel
who
appl
ied
the
full-t
hic
knes
s te
ar)
confirm
ed b
y lo
ngi
tudi
nal
,tr
eatm
ent
and
the
eval
uat
ion
ult
raso
nog
raph
y , o
ver
3 m
onth
s ev
oluti
on.
Clinic
al t
rial
.of
the
resu
lts
was
per
form
ed.
Seve
n e
t al
. (24
)20
1712
0 pa
tien
ts w
ith c
hro
nic
ten
dinop
athy
of
Exp
erim
enta
l,R
ando
miz
ed t
o Blindi
ng
of a
nd
the
eval
uat
ion
IIro
tato
r cu
ff (
tendi
nos
is o
r pa
rtia
l te
ar)
Pro
spec
tive
, 2
trea
tmen
t of
the
resu
lts
was
per
form
ed.
confirm
ed b
y m
agnet
ic r
eson
ance
im
agin
g lo
ngi
tudi
nal
,gr
oups
.Blindi
ng
of t
he
pers
onnel
who
and
ult
raso
nog
raph
y, o
ver
6 m
onth
s C
linic
al t
rial
.ap
plie
d th
e tr
eatm
ent
and
ofev
oluti
on.
the
part
icip
ants
is
not
pos
sibl
e gi
ven t
he
differ
ence
bet
wee
n t
he
type
s of
inte
rven
tion
.C
ole
et a
l. (2
5)20
1836
pat
ients
wit
h c
hro
nic
ten
dinop
athy
of
Exp
erim
enta
l,R
ando
miz
ed t
o Blindi
ng
of p
arti
cipa
nts
and
the
IIth
e su
pras
pinat
us
(ten
dinos
is o
r pa
rtia
l Pro
spec
tive
, 2
trea
tmen
t ev
aluat
ion o
f th
e re
sult
s w
aste
ar <
50%
) co
nfirm
ed b
y ult
raso
und,
ove
r lo
ngi
tudi
nal
,gr
oups
.pe
rfor
med
.3
mon
ths
of e
volu
tion
.C
linic
al t
rial
.Lin
et
al. (
26)
2018
31 p
atie
nts
wit
h c
hro
nic
supr
aspi
nat
us
Exp
erim
enta
l,R
ando
miz
ed t
o Blindi
ng
of p
arti
cipa
nts
, I
tendi
nop
athy
(ten
dinos
is o
r pa
rtia
l te
ar),
Pro
spec
tive
, 2
trea
tmen
t pe
rson
nel
who
appl
ied
the
confirm
ed b
y ult
raso
und,
ove
r 6
mon
ths
longi
tudi
nal
,gr
oups
.tr
eatm
ent
and
the
eval
uat
ion o
f ev
oluti
on.
Clinic
al t
rial
.th
e re
sult
s w
as p
erfo
rmed
.Sa
ri e
t al
. (2
7)20
1912
0 pa
tien
ts w
ith c
hro
nic
ten
dinop
athy
of
Exp
erim
enta
l,R
ando
miz
ed t
o Blindi
ng
of p
arti
cipa
nts
and
IIro
tato
r cu
ff (
tendi
nos
is o
r gr
ade
I pa
rtia
l Pro
spec
tive
, 4
trea
tmen
t pe
rson
nel
who
appl
ied
the
tear
), c
onfirm
ed b
y m
agnet
ic r
eson
ance
lo
ngi
tudi
nal
,gr
oups
.tr
eatm
ent.
imag
ing,
gre
ater
than
3 m
onth
s of
evo
luti
on.
Clinic
al t
rial
.G
eorg
e et
al.
(28)
2018
12 p
atie
nts
wit
h t
endi
nop
athy
of
Exp
erim
enta
l,R
ando
miz
ed t
o It
was
not
rep
orte
d w
het
her
the
III
supr
aspi
nat
us
tendi
nos
is o
r pa
rtia
l te
ar)
Pro
spec
tive
, 2
trea
tmen
t Blindi
ng
of p
arti
cipa
nts
, per
sonnel
co
nfirm
ed w
ith u
ltra
sound,
wit
h s
ympt
oms
longi
tudi
nal
,gr
oups
.an
d th
e ev
aluat
ion o
f th
e re
sult
s gr
eate
r th
an 6
mon
ths
of e
volu
tion
.C
linic
al t
rial
.w
as p
erfo
rmed
.Lee
et
al. (
29)
2015
110
pati
ents
wit
h c
hro
nic
ten
dinop
athy
of
Obs
erva
tion
al,
Not
ran
dom
ized
.N
ot B
lindi
ng.
III
rota
tor
cuff (
tendi
nos
is, p
arti
al t
ear
or
Ret
rosp
ecti
vefu
ll-t
hic
knes
s te
ar),
con
firm
ed b
y C
ase-
Con
trol
s ult
raso
nog
raph
y or
mag
net
ic r
eson
ance
st
udy
.im
agin
g, o
ver
3 m
onth
s ev
oluti
on.
Treb
inja
c et
al.
(30)
2015
21 p
atie
nts
wit
h c
hro
nic
ten
dinop
athy
Obs
erva
tion
alN
ot r
ando
miz
ed.
Not
Blindi
ng.
IV(t
endi
nos
is, p
arti
al t
ear
or full-t
hic
knes
s R
etro
spec
tive
tear
) of
rot
ator
cuff, c
onfirm
ed b
y m
agnet
ic
Cas
e re
cord
s re
sona
nce
imag
ing,
ove
r 6
mon
ths
evol
utio
n.st
udy
.
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
161
AriAs-VAzquez Pi et Al
fIGURe 2. Risk of bias graph of the clinical trials included in the systematic review
fIGURe 3. Risk of bias assessment of the observational studiesincluded in the systematic review
when compared with exercise programs. Similarly, inthe study by Cole et al.25 a mean difference was foundin favor of the HDI group when compared with corti-costeroid infiltrations. On the other hand, in the studyby Sari et al.27, the mean difference was in favor of cor-ticosteroid infiltrations (Figure 4A).
Medium–term follow-up: In the pooled analysis, nostatistically significant difference in pain reduction wasfound when comparing HDI vs. controls (d = -0.009,95% CI -0.448 to 0.430, p(z)>0.968, I2=80.32). In in-dividual analyses, responses similar to those found inshort–terms were observed. In the Seven et al.24 andCole et al.25 studies, the mean differences were foundin favor of the HDI groups, while in the study by Sariet al.27 the observations were in favor of the group whoreceived corticosteroid infiltrations (Figure 4B).
Long–term follow-up: In the pooled analysis, the re-sults showed that HDI had a significant effect on re-ducing pain in individuals with rotator cuff pathology(d = -2.810, 95% CI -4.468 to -1.153, p <0.001,I2=97.86). In individual analyses, the mean differencewas found in favor of the groups that used HDI in thestudies of Seven et al.24 and Bertrand et al.23. None of thestudies found a mean difference that favored the con-trol groups (Figure 4C).
95% CI -0.712 to 0.622, p (z)>0.896, I2=89.22). How-ever, in individual analyses, statistically significant dif-ferences were found: in the study by Seven et al.24, amean difference was found in favor of the HDI group
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
162
HyPertonic dextrose infiltrAtion in rotAtor cuff tendinoPAtHyTA
BLe
II. C
HA
RA
CT
eR
IST
ICS
Of
IN
Te
Rv
eN
TIO
NS
, e
vA
LU
AT
ION
S,
Re
SU
LT
S A
ND
SID
e e
ff
eC
TS
Stu
dy
Inte
rven
tion
Eva
luat
ion
s an
d r
esu
lts
Sid
e ef
fect
sBer
tran
d et
al.
(23)
DX
GR
OU
P: 2
7 pa
tien
ts (
aver
age
age
53.8
yea
rs).
T
he
prim
ary
eval
uat
ions
wer
e pa
in in t
he
shou
lder
Pai
n w
as r
epor
ted
duri
ng
Trea
ted
wit
h 3
ses
sion
s of
mult
i in
ject
ions
wit
h 2
5%
wit
h t
he
VA
S at
the
begi
nnin
g, 1
2 an
d 36
wee
ksan
d af
ter
inje
ctio
n a
s de
xtro
se t
o th
e in
sert
ion o
f te
ndo
ns
of t
he
rota
tor
of fol
low
-up.
only
sid
e ef
fect
s or
cu
ff, a
t m
onth
ly inte
rval
s +
exer
cise
pro
gram
.ad
vers
e re
acti
ons.
ALD
GR
OU
P: 2
0 pa
tien
ts, a
vera
ge a
ge 5
1.1
year
s,
trea
ted
wit
h 3
ses
sion
s of
inje
ctio
ns
wit
h
lido
cain
e/so
luti
on s
alin
e in
the
inse
rtio
n o
f te
ndo
ns
rota
tor
han
dle
+ ex
erci
se p
rogr
am.
ALS
GR
OU
P: 2
7 pa
tien
ts, a
vera
ge a
ge 4
9 ye
ars,
tr
eate
d w
ith 3
ses
sion
s of
subc
uta
neo
us
inje
ctio
ns
wit
h lid
ocai
ne/
solu
tion
sal
ine
+ ex
erci
se p
rogr
am.
Seve
n e
t al
. (24
)D
X G
RO
UP: 6
0 pa
tien
ts, a
vera
ge a
ge 5
0.1
year
s,
They
eva
luat
ed p
ain in t
he
shou
lder
mea
sure
d w
ith
Min
or a
dver
se r
eact
ions
trea
ted
wit
h 6
ses
sion
s of
mult
i in
ject
ions
wit
h
VA
S an
d SP
AD
I at
the
begi
nnin
g, 6
, 12,
48
wee
kssu
ch a
s pa
in,
hyp
erto
nic
dex
tros
e to
the
inse
rtio
n o
f te
ndo
ns
of t
he
of fol
low
-up.
inflam
mat
ion
rota
tor
cuff, u
nde
r ult
raso
und
guid
ance
+ p
roto
col of
an
d hyp
oten
sion
ex
erci
se a
t hom
e.oc
curr
ed.
EX
GR
OU
P: 6
0 pa
tien
ts, a
vera
ge a
ge 4
6.3
year
s,
No
seve
re a
dver
setr
eate
d w
ith e
xerc
ise
prog
ram
.re
acti
ons
wer
e re
port
ed
in a
ny
pati
ent.
Col
e et
al.
(25)
DX
GR
OU
P: 1
7 pa
tien
ts, a
vera
ge a
ge 5
1 ye
ars,
tre
ated
Pai
n w
as a
sses
sed
when
per
form
ing
acti
viti
es a
bove
It
is
not
men
tion
ed if
wit
h a
n inje
ctio
n in t
he
hyp
oech
oic
or a
nec
hoi
c ar
eas
the
hea
d w
ith a
num
eric
al s
cale
fro
m 0
to
4, a
t 6,
12th
ere
wer
e si
de e
ffec
ts o
r of
the
supr
aspi
nat
us
tendo
n (
one
inje
ctio
n),
unde
r an
d 24
wee
ks o
f fo
llow
-up.
adve
rse
reac
tion
s in
ult
raso
und
guid
ance
, at
a ra
te o
f 0.
5ml pe
r zo
ne,
of
pati
ents
.a
25%
dex
tros
e so
luti
on +
exe
rcis
e pr
ogra
m.
CT
GR
OU
P :
19 p
atie
nts
, ave
rage
age
46
year
s, t
reat
ed
wit
h a
n infilt
rati
on in t
he
suba
crom
ial bu
rsa
adja
cent
to t
he
supr
aspi
nat
us
tendo
n, u
nde
r ult
raso
und
guid
ance
, of 2m
l of
a c
ombi
nat
ion o
f 40
mg
of
met
hyl
pred
nis
olon
e ac
etat
e an
d 1%
lid
ocai
ne
+ ex
erci
se p
rogr
am.
cont
inue
s on
the
nex
t pa
ge
VA
S D
X A
LD
A
LS
Bas
al7.
3 (0
.4)
6.9
(0.5
)6.
9(0.
4)
12 w
eeks
-3
.0(0
.51)
-2.7
(0.7
)-
2.7(
0.6)
36
wee
ks-2
.9(0
.6)
-1.8
(0.7
)-1
.3(0
.6)
VA
S D
XE
XBas
al7.
85 (
1.29
)7.
36 (
1.38
)6
wee
ks3.
35 (
1.67
)4.
39(1
.92)
12 w
eeks
2.35
(1.
98)
4.00
(2.
11)
48 w
eeks
0.89
(1.
64)
3.77
(2.1
2)
SPA
DI
DX
EX
Bas
al74
.76(
18.5
4)68
.62
(20.
4)
6 w
eeks
31.3
0(14
.19)
41.9
7(16
.42)
12 w
eeks
16.1
2(12
.82)
37.2
5(20
.32)
48 w
eeks
7.66
(10.
68)
34.9
4(19
.14)
PAIN
DX
CT
Bas
al2.
3(0.
2)2.
6(0.
2)6
wee
ks
2.1
(0.2
)2.
4(0.
2)12
wee
ks.9
(0.2
)2.
1 (0
.3)
24 w
eeks
1.7(
0.2)
1.7(
0.3)
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
163
AriAs-VAzquez Pi et AlTA
BLe
II. C
ON
TIN
UA
TIO
N
Stu
dy
Inte
rven
tion
Eva
luat
ion
s an
d r
esu
lts
Sid
e ef
fect
sLin
et
al. (
26)
DX
GR
OU
P: 1
6 pa
tien
ts, a
vera
ge a
ge 4
6.2
year
s,
Pai
n w
as e
valu
ated
wit
h V
AS
and
SPA
DI, a
t th
e M
inor
adv
erse
rea
ctio
ns
trea
ted
wit
h a
n inje
ctio
n o
f 5m
l of
20%
dex
tros
e in
be
ginnin
g, 2
and
6 w
eeks
of fo
llow
-up.
occu
rred
.th
e ar
ea o
f su
pras
pinat
us
inse
rtio
n, u
nde
r ult
raso
und
No
seve
re a
dver
se
guid
ance
, + e
xerc
ise
prog
ram
.re
acti
ons
wer
e re
port
ed
SS G
RO
UP: 1
5 pa
tien
ts, a
vera
ge a
ge 4
8.6
year
s,
in a
ny
pati
ent.
trea
ted
wit
h a
n infilt
rati
on o
f sa
line
solu
tion
(SS
) in
th
e ar
eas
of s
upr
aspi
nat
us
inse
rtio
n, u
nde
r ult
raso
und
guid
ance
, + e
xerc
ise
prog
ram
.
Sari
et
al. (
27)
DX
GR
OU
P: 3
0 pa
tien
ts, a
vera
ge a
ge 5
2.1
year
s,
Pai
n a
nd
Funct
ional
ity
was
eva
luat
ed w
ith V
AS
and
It is
not
men
tion
ed if
trea
ted
wit
h o
ne
suba
crom
ial in
ject
ion w
ith 5
ml.
of
ASE
S sc
ale
at t
he
begi
nnin
g an
d at
3,1
2,24
wee
ksth
ere
wer
e si
de e
ffec
ts o
r 20
% d
extr
ose
unde
r ult
raso
und
guid
ance
+ p
roto
col
of fol
low
-up:
adve
rse
reac
tion
s in
the
of e
xerc
ise.
pati
ents
tre
ated
.PR
P G
RO
UP: 3
0 pa
tien
ts, a
vera
ge a
ge 5
2.1
year
s,
trea
ted
wit
h o
ne
suba
crom
ial in
ject
ion w
ith 5
ml.
of
plat
elet
-ric
h p
lasm
a unde
r ult
raso
und
guid
ance
+
prot
ocol
of ex
erci
se.
CT
GR
OU
P: 3
0 pa
tien
ts, a
vera
ge a
ge 5
2.1
year
s,
trea
ted
wit
h o
ne
suba
crom
ial in
ject
ion w
ith 5
ml.
of
tria
mci
nol
one
acet
ónid
e (4
0mg)
unde
r ult
raso
und
guid
ance
+ p
roto
col of
exe
rcis
e.A
L G
RO
UP: 3
0 pa
tien
ts, a
vera
ge a
ge 5
2.1
year
s,
trea
ted
wit
h o
ne
suba
crom
ial in
ject
ion w
ith 5
ml.
of
lido
cain
e an
d so
luti
on s
alin
e unde
r ult
raso
und
guid
ance
+ p
roto
col of
exe
rcis
e at
hom
e.
cont
inue
s on
the
nex
t pa
ge
PAIN
D
XSS
Bas
al5.
56 (
0.81
)5.
33 (
0.82
) 2
wee
ks4.
63 (
0.62
)5.
21 (
0.69
)6
wee
ks5.
13(0
.72)
4.87
(0.6
4)
VA
S D
XSS
Bas
al60
.50(
7.87
)65
.00
(2.6
8)
2 w
eeks
52.6
9 (1
0.05
)62
.27(
12.8
2)
6 w
eeks
61.5
6 (4
.58)
60.0
0(4.
9)
VA
SD
XPR
PC
TA
LBas
al5.
95.
635.
635.
47(.
88)
(1)
(.93
)(.
86)
3 w
eeks
4.37
4.83
2.43
4.23
(1.1
6)(.
95)
(1.8
1)(1
.48)
12 w
eeks
4.27
3.9
3.53
3.87
(1.3
6)(.
99)
(1.4
1)(0
.97)
24 w
eeks
3.1
2.57
3.77
3.2
(1.5
2)(1
.19)
(1.4
1)(1
.19)
ASE
SD
XPR
PC
TA
LBas
al45
46.2
40.1
47.2
(9.4
)(8
.6)
(8.8
)(7
.4)
3 w
eeks
52.4
46.1
60.7
55.6
(11.
2)(7
.9)
(11.
4)(1
0.5)
12 w
eeks
56
.155
.758
.158
.8(9
.6)
(7.9
)(9
.03)
(8.8
)24
wee
ks60
.363
.855
.660
.2(1
1.4)
(11.
9)(1
1)(1
1.9)
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
164
HyPertonic dextrose infiltrAtion in rotAtor cuff tendinoPAtHy
TA
BLe
II. C
ON
TIN
UA
TIO
N
Stu
dy
Inte
rven
tion
Eva
luat
ion
s an
d r
esu
lts
Sid
e ef
fect
sG
eorg
e et
al.
(28)
DX
gro
up:
7 p
atie
nts
(m
ean a
ge 6
0 ye
ars)
tre
ated
wit
h
Pai
n a
nd
Funct
ional
ity
was
eva
luat
ed w
ith s
ubs
core
It
is
not
men
tion
ed if
one
dext
rose
inje
ctio
n (
at 1
2.5%
), just
in t
he
foca
l ar
eapa
in D
ASH
and
tota
l sc
ore
resp
ecti
vely
, at
ther
e w
ere
side
effec
ts o
r of
ten
dinos
is o
r ru
pture
unde
r ult
raso
und
guid
ance
.th
e be
ginnin
g an
d at
12
wee
ks o
f fo
llow
-up:
adve
rse
reac
tion
s in
the
PH
GR
OU
P: 5
pat
ients
(m
ean a
ge 5
8 ye
ars)
tre
ated
tr
eate
d pa
tien
ts.
wit
h c
onve
nti
onal
phys
ioth
erap
y.
Lee
et
al. (
29)
DX
gro
up:
57
pati
ents
; ave
rage
age
of 54
.1 y
ears
, Pai
n w
as e
valu
ated
wit
h V
AS
and
funct
ional
ity
wit
h
It is
not
men
tion
ed if
trea
ted
wit
h 4
(3
- 8)
dex
tros
e in
filt
rati
ons
(10m
l. at
SP
AD
I, b
efor
e tr
eatm
ent
and
48 w
eeks
aft
er t
he
ther
e w
ere
side
effec
ts o
r 16
.5%
) in
supr
aspi
nat
us
and
subs
capu
lar
tendo
ns,
at
appl
icat
ion o
f th
e sa
me.
adve
rse
reac
tion
s in
the
inte
rval
s ev
ery
2 -
4 w
eeks
.tr
eate
d pa
tien
ts.
CL G
roup:
53
pati
ents
, ave
rage
age
of 55
.8 y
ears
who
conti
nued
wit
h t
he
sam
e co
nse
rvat
iv t
reat
men
t pr
evio
usl
y es
tabl
ished
.
Treb
inja
c et
al.
(30)
21 p
atie
nts
; ave
rage
age
of 47
.8 y
ears
, tre
ated
wit
h
Pai
n w
as e
valu
ated
wit
h V
AS
and
funct
ional
ity
wit
h
It is
not
men
tion
ed if
6 in
tra
– ar
ticu
lar
dext
rose
inje
ctio
ns,
(6
ml at
25%
) SP
AD
I be
fore
tre
atm
ent
and
48 w
eeks
aft
er t
he
ther
e w
ere
side
effec
ts o
r an
d ex
tra
- ar
ticu
lar
(1 m
l pe
r de
xtro
se p
oint
at 1
5%),
ap
plic
atio
n o
f th
e sa
me.
adve
rse
reac
tion
s in
the
at m
onth
ly inte
rval
s.tr
eate
d pa
tien
ts.
VA
S
DX
CL
Bas
al6.
3 (1
.0)
6.1
(1.2
)
48 w
eeks
2.7(
1.0)
4.6(
1.4)
SPA
DI
DX
CL
Bas
al69
.4(9
.2)
67.6
(9.4
)
48 w
eeks
43.8
(11
.6)
51.1
(14.
4)
DA
SH s
ubs
core
Pai
n
DX
PH
Bas
al3.
293.
20
12 w
eeks
1.86
2.40
DA
SH t
otal
sco
re
DX
PH
Bas
al60
.14
56.8
6
12 w
eeks
43.8
946
.68
VA
SSP
AD
I
Bas
al8.
14 (
1.2)
76.9
9 (1
3.6)
48 w
eeks
2.29
(2.8
)20
.84
(23.
06)
Abb
revi
atio
ns:
ALD
= d
eep
loca
l anes
thet
ics,
ALS=
subc
uta
neo
us
loca
l anes
thet
ics,
DX
: dex
tros
e, E
X=
Exe
rcis
e, S
S= s
oluti
on s
alin
e, C
T=
cort
icos
tero
id, A
L=
loca
l anes
thet
ics,
PR
P=
plat
elet
-ric
h p
lasm
a, P
H=
Phys
ioth
erap
y, C
L=
Con
trol
. SPA
DI =
Shou
lder
Pai
n a
nd
Dis
abilit
y In
dex,
VA
S=A
nal
og V
isual
Sca
le. A
SES=
Am
eric
an S
hou
lder
and
Elb
owSu
rgeo
ns
Stan
dard
ized
, DA
SH: D
isab
ilit
y of
Arm
and
Shou
lder
Sco
re.
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
165
AriAs-VAzquez Pi et Al
RevIew Of OBSeRvATIONAL STUDIeS
In the observational studies, Lee et al.29 reported sta-tistically significant reduction of pain and improvementof function in the HDI group compared with conser-vative treatments in long-term follow-up. Similarly, Tre-binjac et al.30 reported a series of cases treated with HDIwith statically significant pain reduction and functionimprovement in the long-term.
CHARACTeRISTICS AND DOSAGe Of HypeRTONIC
DexTROSe TReATMeNT
Four studies23,24,29,30 performed treatment schemes withmultiple sessions and multi-injections in the insertionof the rotator cuff tendons; other studies25,26,28 only useda single intratendinous application in the focal area ofrupture under ultrasound guidance. Sari et al.27 per-formed a single subacromial infiltration.
In the included studies, the number of sessions va -ried from 1 to 8 per participant, while the application
frequency was every 2 to 4 weeks. On the other hand,the concentrations of dextrose used varied from 12.5to 25%, with a mode of 25%.
ADveRSe ReACTIONS AND SIDe effeCTS
Three trials23,24,26 where participants received HDI re-ported minor adverse reactions such as pain during orafter application, inflammation after application andone study reported hypotension during treatment. Therest of the studies did not mention if there were ad-verse reactions.
DISCUSSION
effICACy Of INfILTRATIONS wITH HypeRTONIC
DexTROSe IN THe TReATMeNT Of ROTATOR CUff
TeNDINOpATHy
A recent review already evaluated the role of HDI in
fIGURe 4. Forest plot: A) Short-term, B) Medium-term, C) Long-term.
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
166
HyPertonic dextrose infiltrAtion in rotAtor cuff tendinoPAtHy
TA
BLe
III
. S
UM
MA
Ry
Of
TH
e C
HA
RA
CT
eR
IST
ICS
AN
D p
RO
pe
RT
IeS
Of
TH
e I
Nf
ILT
RA
TIO
NS
US
eD
fO
R T
He
TR
eA
TM
eN
T O
f p
AT
IeN
TS
wIT
H
RO
TA
TO
R C
Uf
f T
eN
DIN
Op
AT
Hy
IN
TH
e S
TU
DIe
S I
NC
LU
De
D I
N T
He
Re
vIe
w.
Mec
han
ism
E
ffec
ts o
n
Tre
atm
ent
mod
alit
ies
of a
ctio
np
ain
con
trol
Ad
vers
e ef
fect
sP
ossi
ble
adva
nta
ges
Pos
sibl
e d
isad
van
tage
sC
orti
cost
eroi
d in
filt
rati
onA
nti
-inflam
mat
ory
effe
ct.
Shor
t te
rm.
Pai
n d
uri
ng
appl
icat
ion.
Usu
ally
the
appl
icat
ion
Its
appl
icat
ion is
lim
ited
Lin
MT
et
al. (
5)Pro
babl
e de
lete
riou
s sc
hem
e in
volv
es a
sin
gle
to v
ery
few
ses
sion
s to
Col
e et
al.
(25)
effe
cts
to t
he
tendo
n if
inje
ctio
n intr
a-ar
ticu
lar
avoi
d ad
vers
e ef
fect
s.Sa
ri e
t al
. (27
)re
peat
ed infilt
rati
ons
are
or in t
he
suba
crom
ial
Intr
aten
dinou
s C
ook
et a
l. (4
0)ca
rrie
d ou
t.sp
ace.
appl
icat
ion is
not
R
amír
ez e
t al
. (41
)Pos
sibi
lity
of sy
stem
ic
reco
mm
ende
d.Ji
et
al. (
42)
side
effec
ts.
Loc
al a
nes
thet
ic infilt
rati
onIn
hib
itor
of noc
icep
tive
Sh
ort
term
.Pai
n d
uri
ng
appl
icat
ion.
It c
an b
e ap
plie
dPos
sibl
e ef
fect
only
in
Ber
tran
d et
al.
(23)
acti
vity
and
slig
ht
intr
a-ar
ticu
larl
y, in t
he
the
very
shor
t te
rm.
Sari
et
al. (
27)
anti
-inflam
mat
ory
effe
ct.
suba
crom
ial sp
ace
or in t
he
Car
acas
et
al. (
37)
enth
esis
of th
e te
ndo
n.
Coo
k et
al.
(40)
Hyp
erto
nic
dex
tros
e Tr
ophic
effec
t on
the
Lon
g te
rm.
Pai
n d
uri
ng
and
afte
r It
can
be
appl
ied
The
schem
es a
pplied
infilt
rati
onte
ndo
n.
appl
icat
ion.
intr
a-ar
ticu
larl
y, in t
he
invo
lve
mult
iple
Lin
MT
et
al. (
5)su
bacr
omia
l sp
ace,
in t
he
inje
ctio
ns
in t
he
sam
eBer
tran
d et
al.
(23)
enth
esis
of th
e te
ndo
n o
rse
ssio
n.
Seve
n e
t al
. (24
)in
trat
endi
nou
s.C
ole
et a
l. (2
5)M
ult
iple
ses
sion
s ca
n b
e Lin
et
al. (
26)
appl
ied,
whic
h c
ould
be
Sari
et
al. (
27)
use
ful in
ref
ract
ory
case
s.G
eorg
e et
al.
(28)
Lee
et
al. (
29)
Treb
inja
c et
al.
(30)
Cat
apan
o et
al.
(31)
Kin
et
al. (
38)
Ahn e
t al
. (39
)Pel
t et
al.
(45)
cont
inue
s on
the
nex
t pa
ge
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
167
AriAs-VAzquez Pi et Al
TA
BLe
III
. C
ON
TIN
UA
TIO
N
Mec
han
ism
E
ffec
ts o
n
Tre
atm
ent
mod
alit
ies
of a
ctio
np
ain
con
trol
Ad
vers
e ef
fect
sP
ossi
ble
adva
nta
ges
Pos
sibl
e d
isad
van
tage
sPla
tele
t-ri
ch p
lasm
a Tr
ophic
effec
t on
the
Lon
g te
rm.
Pai
n d
uri
ng
and
afte
r It
can
be
appl
ied
Com
plex
ity
and
infilt
rati
onte
ndo
n.
appl
icat
ion.
intr
a-ar
ticu
larl
y, in t
he
vari
abilit
y in
Lin
MT
et
al. (
5)su
bacr
omia
l sp
ace
orpr
epar
atio
n t
echniq
ue.
Chen
et
al. (
9)in
trat
endi
nou
s.Sa
ri e
t al
. (27
)Per
hap
s it
is
the
infilt
rati
on
Pau
ly e
t al
. (43
)m
odal
ity
wit
h t
he
grea
test
D
hura
t et
al.
(44)
rege
ner
ativ
e po
tenti
al.
the management of rotator cuff tendinopathy; howev-er, it only performed qualitative analyses31. In our re-view, more recent studies were added and we addi-tionally performed quantitative analyses to clarify theefficacy of HDI.
In the pooled analysis, we found that HDI were moreeffective for reducing pain than others treatments usedin controls in a long-term follow-up. This observationis similar to the results of the meta-analysis by Lin etal.5, who reported that HDI are an effective interventionfor long-term pain control in individuals with rotatorcuff tendinopathy (nevertheless, that meta-analysisonly included one study where HDI were used). Thesame long-term symptomatic effects were reported inthe observational studies29,30.
The study by Seven et al.24 compared HDI vs. exer-cise programs, and found that HDI was more effectivein reducing pain in short, medium and long-terms in theindividual analysis. Non-invasive therapeutic strategiesare widely used for the treatment of rotator cufftendinopathy32. It has been reported that oral anti-in-flammatory drugs are effective for reducing pain only inthe short term, but they do not improve function2. Like-wise, physiotherapy modalities are frequently used forthe treatment of rotator cuff tendinopathy; neverthe-less, it has been reported that some of them such as thetranscutaneous electrical stimulation33 and therapeuticultrasound34 are not more effective than placebo for thecontrol of pain and improvement of function, while oth-ers such as laser therapy have shown a small beneficialeffect3,35. Therapeutic exercise probably represents themost effective non-invasive therapeutic modality for thetreatment of rotator cuff tendinopathy4,36. No previousmeta-analysis has reported a direct comparison betweenHDI vs. exercise programs; the results of our meta-anal-ysis suggest that HDI are more effective in the short,medium and long terms than exercise programs andcould be an alternative when exercise strategies fail.However, this comparison was only carried out in onestudy, so it should be taken with reserve.
Some studies directly compared HDI vs. infiltrationswith local anesthetics23,27. In the individual analyses inthe short and medium terms, no significant differenceswere reported between the groups. However, in thelong term however, the individual analysis of the studyby Bertrand et-al. showed significant differences in fa-vor of the group treated with HDI23. A previous meta-analysis5 reported that HDI are an effective interven-tion for the long-term control of pain in individualswith rotator cuff tendinopathy, based on the compari-
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
168
HyPertonic dextrose infiltrAtion in rotAtor cuff tendinoPAtHy
son between HDI and local anesthetic infiltrations.These results suggest that in the short and mediumterms HDI and local anesthetics infiltrations have thesame effect in patients with rotator cuff tendinopathy.However, in the long term, the effect of local anesthet-ics is lost, while the benefit achieved with HDI persists.Local anesthetics have been proposed to have analgesicand anti-inflammatory effects in addition to their anes-thetic effect37, which could explain its short-term ther-apeutic effect. Hypertonic dextrose on the other hand,has been proposed to have a mechanism of action basedon the increase of fibroblast proliferation, collagen pro-duction and extracellular matrix in the treated ten-dons38,39, which could explain why its efficacy is main-tained in the long-term.
Two studies directly compared HDI vs. conventio nalinfiltrations with corticosteroids25, 27. In the individualanalyses, the study of Cole et al.25 reported statisticallysignificant pain reduction in the short and mediumterms in favor of HDI, but not in the long term. On theother hand, the study by Sari et al.27 reported a statis-tically significant reduction of pain in favor of corti-costeroid infiltrations in the short and medium terms,but in the long-term there was no difference betweengroups. No previous meta-analysis has reported a di-rect comparison between HDI vs. corticosteroid infil-trations. Other meta-analyses have reported that corti-costeroid infiltrations are an effective intervention forpain control and function improvement in the rotatorcuff tendinopathy when compared with placebo5 or lo-cal anesthetics40; however, it was observed that the im-provement only lasted <6 weeks. Corticosteroids areprobably the most used infiltration in individuals withshoulder rotator cuff tendinopathy but its use has beenrelated to deleterious effects on the tendon in additionto its contraindication in some patient with comor-bidities. Basic studies have reported that infiltrationswith corticosteroids could be associated with an in-crease of cellular apoptosis in the infiltrated tendon41
and could facilitate the NF-KB signaling, which is in-volved in the pathogenesis of rotator cuff tears42, whichcontrasts with the trophic effects that hyperto nic dex-trose can have on the tendon38,39. When compa r ing HDIvs. corticosteroids infiltrations in individuals with ro-tator cuff tendinopathy, further studies are required toclarify whether HDI represents an alternative to infil-trations with corticosteroids when there is a con-traindication for their application.
Sari et al.27 compared HDI vs. infiltrations with PRPand in the individual analyses found no significant
diffe rences between both groups in the short, mediumand long terms. No previous meta-analysis reported adirect comparison between HDI vs. PRP infiltration.Others meta-analysis reported that PRP infiltrationswere effective for reducing pain and improving func-tion in long-term follow-up when compared with cor-ticosteroids9. In basic studies it has been reported thatin vitro applications of PRP favor tissue repair and in-crease the proliferation of rotator cuff tenocytes43, sim-ilar to the effects previously described for hypertonicdextrose38, 39; nonetheless, the application of PRP im-plies greater complexity in its preparation and the re-sults vary considerably44. Although HDI appear to havethe same efficacy as PRP infiltrations to reduce pain inthe long-term, these observations come from a singlestudy and more studies are necessary to corroboratethese results.
Regarding the characteristics of the treatment, stud-ies with multiple sessions and multi-injections23,24,29,30
showed greater benefits with statistically significant im-provement in favor of the groups treated with HDI. Onthe other hand, studies25, 28 that only included one treat-ment session (intratendinous) did not find greater clin-ical improvement in comparison with control groups.Multiple sessions and multi-injection schemes appearto be necessary to obtain clinical benefit in patientswith rotator cuff tendinopathy. Previous recommenda-tions suggest between 3 to 6 treatment sessions as wellas multi-injection schemes involving at least the inser-tion area of the rotator cuff tendons and subacromial orintra-articular space45.
Regarding the side effects and/or adverse reactions,pain during or after the application was the most fre-quently observed. None of the treated individuals re-ported serious complications such as infections or al-lergic reactions; nevertheless, not all studies reportedadverse effects or complications.
A comparative table was made, which summarizesthe mechanisms of action, effects on pain control, pos-sible advantages, disadvantages and adverse effects, ofthe types of infiltration used for the treatment of rota-tor cuff tendinopathy in the included studies (Table III).
It is important to mention possible limitations of thepresent study. Although the studies included in thequantitative analysis showed a good methodologicalquality design and low or moderate risk of bias, thesmall number of studies included, the small number ofindividuals treated in each study and the lack of stan-dardization in the application techniques most have in-fluenced the results and its consistency.
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
169
AriAs-VAzquez Pi et Al
CONCLUSIONS
The results of this systematic review and meta-analysisindicate that HDI are an effective treatment for controlpain in long-term follow-up of individuals with rotatorcuff tendinopathy. Therefore, it can be concluded thatHDI were more effective than non-invasive treatmentsin the short, medium and long terms, as well as moreeffective than the use of local anesthetics in long term.HDI were not more effective than PRP and its efficacycomparing with corticosteroid infiltrations is not yetclear. On the other hand, HDI did not show complica-tions or serious adverse effect. Despite the favorable re-sults, the small number of studies included in our meta-analysis as well as their heterogeneity are the mainlimitations to draw definitive conclusions and goodquality RCTs are required.
CORReSpONDeNCe TO
Carlos Alfonso Tovilla-ZárateUniversidad Juárez Autónoma de Tabasco División Académica Multidisciplinaria de Comalcalco, RancheríaSur, Cuarta Sección C.P. 86650, Comalcalco, Tabasco, México. E-mail: [email protected]
RefeReNCeS
1. Tashjian RZ. Epidemiology, natural history, and indications fortreatment of rotator cuff tears. Clin Sports Med 2012; 31: 589--604.
2. Boudreault J, Desmeules F, Roy JS, Dionne C, Frémont P,Mac-dermid JC. The efficacy of oral non-steroidal anti-inflammato-ry drugs for rotator cuff tendinopathy: a systematic review andmeta-analysis. J Rehabil Med 2014; 46: 294-306.
3. Page MJ, Green S, Mrocki MA, Surace SJ, Deitch J, McBain B, etal. Electrotherapy modalities for rotator cuff disease. CochraneDatabase Syst Rev 2016; Cd012225.
4. Hanratty CE, McVeigh JG, Kerr DP, Basford JR, Finch MB,Pendleton A, et al. The effectiveness of physiotherapy exercisesin subacromial impingement syndrome: a systematic review andmeta-analysis. Semin Arthritis Rheum 2012; 42: 297-316.
5. Lin MT, Chiang CF, Wu CH, Huang YT, Tu YK,Wang TG. Com-parative Effectiveness of Injection Therapies in Rotator CuffTendinopathy: A Systematic Review, Pairwise and NetworkMeta-analysis of Randomized Controlled Trials. Arch Phys MedRehabil 2019; 100: 336-349.e15.
6. Lin KM, Wang D,Dines JS. Injection Therapies for Rotator CuffDisease. Orthop Clin North Am 2018; 49: 231-239.
7. Tashjian RZ. AAOS clinical practice guideline: optimizing themanagement of rotator cuff problems. J Am Acad Orthop Surg2011; 19: 380-3.
8. Pandey V,Jaap Willems W. Rotator cuff tear: A detailed update.Asia Pac J Sports Med Arthrosc Rehabil Technol 2015; 2: 1-14.
9. Chen X, Jones IA, Park C,Vangsness CT, Jr. The Efficacy ofPlatelet-Rich Plasma on Tendon and Ligament Healing: A Sys-tematic Review and Meta-analysis With Bias Assessment. Am JSports Med 2018; 46: 2020-2032.
10. Osti L, Buda M, Buono AD, Osti R,Massari L. Clinical evidencein the treatment of rotator cuff tears with hyaluronic acid. Mus-cles Ligaments Tendons J 2015; 5: 270-5.
11. Kim SJ, Kim EK, Kim SJ,Song DH. Effects of bone marrow aspi-rate concentrate and platelet-rich plasma on patients with partialtear of the rotator cuff tendon. J Orthop Surg Res 2018; 13: 1.
12. Ikonomidis S, Iliakis E,Charalambus Dvakirtzian L. Nonopera -tive treatment of shoulder impingement syndrome with topicalinjections of medical oxygen-ozone mixture. A double blindclinical trial. Rivista Italiana di Ossigeno-Ozonoterapia 2002; 1:41-44.
13. Hauser RA, Lackner JB, Steilen-Matias D,Harris DK. A System-atic Review of Dextrose Prolotherapy for Chronic Muscu-loskeletal Pain. Clin Med Insights Arthritis Musculoskelet Dis-ord 2016; 9: 139-159.
14. Rabago D,Nourani B. Prolotherapy for osteoarthritis andtendinopathy: a descriptive review. Current rheumatology re-ports 2017; 19: 34.
15. Arias-Vázquez PI, Tovilla-Zárate CA, Legorreta-Ramírez BG, Bu-rad Fonz W, Magaña-Ricardez D, González-Castro TB, et al. Pro-lotherapy for knee osteoarthritis using hypertonic dextrose vsother interventional treatments: systematic review of clinicaltrials. Adv Rheumatol 2019; 59: 39.
16. Morath O, Kubosch EJ, Taeymans J, Zwingmann J, Konstan-tinidis L, Südkamp NP, et al. The effect of sclerotherapy andprolotherapy on chronic painful Achilles tendinopathy-a sys-tematic review including meta-analysis. Scand J Med Sci Sports2018; 28: 4-15.
17. Neph A, Onishi K,Wang JH. Myths and Facts of In-Office Re-generative Procedures for Tendinopathy. Am J Phys Med Reha-bil 2019; 98: 500-511.
18. Dwivedi S, Sobel AD, DaSilva MF,Akelman E. Utility of Pro-lotherapy for Upper Extremity Pathology. J Hand Surg Am 2019;44: 236-239.
19. Moher D, Liberati A, Tetzlaff J,Altman DG. Preferred reportingitems for systematic reviews and meta-analyses: the PRISMAstatement. PLoS Med 2009; 6: e1000097.
20. Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ,et al., Cochrane handbook for systematic reviews of interven-tions. 2019: John Wiley & Sons.
21. Sterne JA, Hernán MA, Reeves BC, Savovi� J, Berkman ND,Viswanathan M, et al. ROBINS-I: a tool for assessing risk of biasin non-randomised studies of interventions. Bmj 2016; 355:i4919.
22. Sullivan D, Chung KC, Eaves FF, 3rd,Rohrich RJ. The level ofevidence pyramid: indicating levels of evidence in Plastic andReconstructive Surgery articles. Plast Reconstr Surg 2011; 128:311-314.
23. Bertrand H, Reeves KD, Bennett CJ, Bicknell S,Cheng AL. Dex-trose Prolotherapy Versus Control Injections in Painful RotatorCuff Tendinopathy. Arch Phys Med Rehabil 2016; 97: 17-25.
24. Seven MM, Ersen O, Akpancar S, Ozkan H, Turkkan S, YıldızY, et al. Effectiveness of prolotherapy in the treatment of chron-ic rotator cuff lesions. Orthop Traumatol Surg Res 2017; 103:427-433.
25. Cole B, Lam P, Hackett L,Murrell GAC. Ultrasound-guided in-jections for supraspinatus tendinopathy: corticosteroid versusglucose prolotherapy - a randomized controlled clinical trial.Shoulder Elbow 2018; 10: 170-178.
26. Lin CL, Huang CC,Huang SW. Effects of hypertonic dextrose in-jection in chronic supraspinatus tendinopathy of the shoulder:
THE OffiCiAL JOURnAL Of THE PORTUGUESE SOCiETY Of RHEUMATOLOGY
170
HyPertonic dextrose infiltrAtion in rotAtor cuff tendinoPAtHy
a randomized placebo-controlled trial. Eur J Phys Rehabil Med2019; 55: 480-487.
27. Sari A,Eroglu A. Comparison of ultrasound-guided platelet-richplasma, prolotherapy, and corticosteroid injections in rotatorcuff lesions. J Back Musculoskelet Rehabil 2020; 33: 387-396.
28. George J, Li SC, Jaafar Z,Hamid MSA. Comparative Effectivenessof Ultrasound-Guided Intratendinous Prolotherapy Injectionwith Conventional Treatment to Treat Focal SupraspinatusTendinosis. Scientifica (Cairo) 2018; 4384159.
29. Lee DH, Kwack KS, Rah UW,Yoon SH. Prolotherapy for Refrac-tory Rotator Cuff Disease: Retrospective Case-Control Study of1-Year Follow-Up. Arch Phys Med Rehabil 2015; 96: 2027-32.
30. Trebinjac S,Kitchbi H. Long-term effect of Prolotherapy onsymptomatic rotator cuff tendinopathy. J Journal of Health Sci-ences 2015; 5: 93-98.
31. Catapano M, Zhang K, Mittal N, Sangha H, Onishi K,de Sa D.Effectiveness of Dextrose Prolotherapy for Rotator CuffTendinopathy: A Systematic Review. Pm r 2020; 12: 288-300.
32. Dang A,Davies M. Rotator Cuff Disease: Treatment Options andConsiderations. Sports Med Arthrosc Rev 2018; 26: 129-133.
33. Desmeules F, Boudreault J, Roy JS, Dionne CE, Frémont P,Mac-Dermid JC. Efficacy of transcutaneous electrical nerve stimula-tion for rotator cuff tendinopathy: a systematic review. Physio-therapy 2016; 102: 41-49.
34. Desmeules F, Boudreault J, Roy JS, Dionne C, Frémont P,Mac-Dermid JC. The efficacy of therapeutic ultrasound for rotatorcuff tendinopathy: A systematic review and meta-analysis. PhysTher Sport 2015; 16: 276-284.
35. Hawk C, Minkalis AL, Khorsan R, Daniels CJ, Homack D, GliedtJA, et al. Systematic Review of Nondrug, Nonsurgical Treatmentof Shoulder Conditions. J Manipulative Physiol Ther 2017; 40:293-319.
36. Abdulla SY, Southerst D, Côté P, Shearer HM, Sutton D, Rand-
hawa K, et al. Is exercise effective for the management of sub-acromial impingement syndrome and other soft tissue injuriesof the shoulder? A systematic review by the Ontario Protocol forTraffic Injury Management (OPTIMa) Collaboration. Man Ther2015; 20: 646-656.
37. Caracas HC, Maciel JV, Martins PM, de Souza MM,Maia LC. Theuse of lidocaine as an anti-inflammatory substance: a systemat-ic review. J Dent 2009; 37: 93-97.
38. Kim HJ, Jeong TS, Kim WS,Park YS. Comparison of Histologi-cal Changes in Accordance with the Level of Dextrose-Concen-tration in Experimental Prolotherapy Model. Journal of the Ko-rean Academy of Rehabilitation Medicine 2003; 27: 935-940.
39. Ahn KH, Kim HS, Lee WK, Kim HW, Yun DH,Kim DH. The Ef-fect of the Prolotherapy on the Injured Achilles Tendon in a RatModel. Journal of the Korean Academy of RehabilitationMedicine 2002; 26: 332-336.
40. Cook T, Minns Lowe C, Maybury M,Lewis JS. Are corticosteroidinjections more beneficial than anaesthetic injections alone inthe management of rotator cuff-related shoulder pain? A sys-tematic review. Br J Sports Med 2018; 52: 497-504.
41. Ramírez JP, Bonati-Richardson F, García MP, Hidalgo C, StooreC, Liendo R, et al. Intra-articular treatment with corticosteroidsincreases apoptosis in human rotator cuff tears. Connect TissueRes 2019; 60: 283-290.
42. Ji JH, Kim YY, Patel K, Cho N, Park SE, Ko MS, et al. Dexam-ethasone Facilitates NF-�B Signal Pathway in TNF-� StimulatedRotator Cuff Tenocytes. J Microbiol Biotechnol 2019; 29: 297--303.
43. Pauly S, Klatte-Schulz F, Stahnke K, Scheibel M,Wildemann B.The effect of autologous platelet rich plasma on tenocytes of thehuman rotator cuff. BMC Musculoskeletal Disorders 2018; 19:422.
44. Dhurat R,Sukesh M. Principles and Methods of Preparation ofPlatelet-Rich Plasma: A Review and Author's Perspective. J Cu-tan Aesthet Surg 2014; 7: 189-197.
45. Pelt RSV. Shoulder Prolotherapy Injection Technique. Journal ofProlotherapy 2009; 1: 243-245.