981
REVIEW ARTICLE (META-ANALYSIS)
Carpal Tunnel Syndrome. Part I: Effectiveness of Nonsurgical
Treatments–A Systematic Review
Bionka M. Huisstede, PhD, Peter Hoogvliet, MD, PhD, Manon S. Randsdorp, MD, Suzanne Glerum, MD,
Marienke van Middelkoop, PhD, Bart W. Koes, PhD
ABSTRACT. Huisstede BM, Hoogvliet P, Randsdorp MS,
Glerum S, van Middelkoop M, Koes BW. Carpal tunnel syn-
drome. Part I: effectiveness of nonsurgical treatments–a sys-
tematic review. Arch Phys Med Rehabil 2010;91:981-1004.
Objective: To review literature systematically concerning
effectiveness of nonsurgical interventions for treating carpal
tunnel syndrome (CTS).
Data Sources: The Cochrane Library, PubMed, EMBASE,
CINAHL, and PEDro were searched for relevant systematic
reviews and randomized controlled trials (RCTs).
Study Selection: Two reviewers independently applied the
inclusion criteria to select potential studies.
Data Extraction: Two reviewers independently extracted
the data and assessed the methodologic quality.
Data Synthesis: A best-evidence synthesis was performed
to summarize the results of the included studies. Two reviews
and 20 RCTs were included. Strong and moderate evidence
was found for the effectiveness of oral steroids, steroid injec-
tions, ultrasound, electromagnetic field therapy, nocturnal
splinting, and the use of ergonomic keyboards compared with
a standard keyboard, and traditional cupping versus heat pads
in the short term. Also, moderate evidence was found for
ultrasound in the midterm. With the exception of oral and
steroid injections, no long-term results were reported for any of
these treatments. No evidence was found for the effectiveness
of oral steroids in long term. Moreover, although higher doses
of steroid injections seem to be more effective in the midterm,
the benefits of steroids injections were not maintained in the
long term. For all other nonsurgical interventions studied, only
limited or no evidence was found.
Conclusions: The reviewed evidence supports that a number
of nonsurgical interventions benefit CTS in the short term, but
there is sparse evidence on the midterm and long-term effec-
tiveness of these interventions. Therefore, future studies should
concentrate not only on short-term but also on midterm and
long-term results.
Key Words: Carpal tunnel syndrome; Rehabilitation; Re-
view [publication type]; Treatment outcome.
© 2010 by the American Congress of Rehabilitation
Medicine
From Department of General Practice (Huisstede, Randsdorp, Glerum,
van Middelkoop, Koes) and Department of Rehabilitation Medicine (Huisstede,
Hoogvliet, Randsdorp), Erasmus Medical Center, Rotterdam, The Netherlands.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit on the authors or on any organi-
zation with which the authors are associated.
Reprint requests to Bionka M. Huisstede, PhD, Erasmus Medical Center—
University Medical Center Rotterdam, Dept of Rehabilitation Medicine, Room
H-016, PO Box 2040, 3000 CA Rotterdam, The Netherlands, e-mail:
b.huisstede@erasmusmc.nl.
0003-9993/10/9107-00941$36.00/0
doi:10.1016/j.apmr.2010.03.022
ARPAL TUNNEL SYNDROME, 1 of the 6 peripheral
C nerve entrapments located in the upper extremity, is 1
caused by compression of the median nerve as it passes through
the carpal tunnel.
Twenty-nine percent of those with chronic complaints of the
upper extremity reported complaints in the wrist/hand area.2
The prevalence of possible or probable CTS in the general
population depends on nuances of the definition used, but it is
cited as being 5.3% in women and 2.1% in men.3 Among those
with work-related upper-extremity disorders, work-related
CTS is one of the most disabling and costly, representing a
major cause of lost work days and workers’ compensation costs
in the United States (U.S. Department of Health and Human
Services, 1996). In the United States, 400,000 operations to
treat CTS are performed each year, costing a total of $2
billion.4
Characteristic complaints of CTS are pain, paresthesia, and
numbness in the fingers and hand (in the area innervated by the
median nerve), often exacerbated at night.5 The exact patho-
physiology of how the pressure in the carpal tunnel increases
over time is unclear,6 although it is known that the occurrence
of CTS is associated with an average hand force requirement of
greater than 4kg, repetitiveness at work (cycle time ⬍10s, or
⬎50% of cycle time performing the same movements), and a
daily 8-hour energy-equivalent frequency-weighted accelera-
tion of 3.9m/s2.7
Many interventions, both nonsurgical and surgical, have
been suggested to treat CTS. No therapy for CTS is universally
accepted,8 although monodisciplinary as well as multidisci-
plinary clinical guidelines have been developed.9,10
Nonsurgical treatment options vary from rest or activity
modification to splinting, or the use of oral medication such as
nonsteroidal anti-inflammatory drugs or oral steroids.11 In de-
compression surgery, open as well as endoscopic techniques
have been used. The most frequently reported treatments are
splinting (56.3%) and nonsteroidal anti-inflammatory agents
(50.8%).12 Two Cochrane reviews have been written concern-
ing nonsurgical treatment options to treat CTS. One of these
reviews13 concerned the effectiveness of all types of nonsur-
gical treatments other than steroid injections. This review
showed short-term benefit from treatment with ultrasound,
splinting, oral steroids, yoga, and carpal bone mobilization. No
significant results were found in favor of other nonsurgical
List of Abbreviations
CI confidence interval
CTS carpal tunnel syndrome
DASH Disability of the Arm, Shoulder and Hand
MD mean difference
RCT randomized controlled trial
RR relative risk
VAS visual analog scale
WMD weighted mean difference
Arch Phys Med Rehabil Vol 91, July 2010
982 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
treatments. The second review14 reported on the effectiveness
of local corticosteroid injections. Corticosteroid injections
were more effective than placebo after 1 month and also more
effective than oral corticosteroids after 3 months. No signifi-
cant clinical benefit was found for corticosteroid injections
compared with other treatments or in favor of multiple injec-
tions compared with 1 injection.
Since the publication of these Cochrane reviews, several
RCTs have been published, and we wondered whether the
conclusions made in the Cochrane reviews would remain
the same or would need modification. To optimize further the
quality of care for patients with CTS given by clinicians and by
medical and paramedical staff working in primary care, an
overview of the current state of the art regarding evidence-
based information is needed that can support developing and
updating evidence-based protocols and guidelines for interven-
tions. Therefore, we systematically reviewed scientific litera-
ture to provide an up-to-date overview of the evidence for the
effectiveness of interventions to treat CTS. This article, part I,
concentrates on nonsurgical interventions to treat CTS.
METHODS
Search Strategy
A search of relevant systematic reviews on CTS was per-
formed in the Cochrane Library. In addition, relevant reviews
and RCTs in PubMed, EMBASE, CINAHL, and PEDro were
searched (1) for interventions included in the systematic re-
views from the date of the search strategy of the review up to
January 2010 (ie, recent RCTs), and (2) from the beginning of
the database to January 2010 (ie, additional RCTs).
Key words related to the disorder such as “carpal tunnel
syndrome,” “median nerve entrapment,” and “interventions”
were included in the literature search. The complete search
strategy is described in appendix 1.
Inclusion Criteria
Systematic reviews and/or RCTs were considered eligible
for inclusion if they fulfilled all of the following criteria: (1) the
study included patients with CTS, (2) CTS was not caused by
an acute trauma or any systemic disease (such as osteoarthritis,
rheumatoid arthritis, diabetes mellitus, or other connective
tissue disease) as described in the definition of complaints of
the arm, neck, and/or shoulder (CANS), (3) an intervention for
treating the disorder was evaluated, and (4) results on pain,
function or recovery were reported. There were no language
restrictions.
If a subset of the total number of patients included in a study
met our inclusion criteria, the study was included only if the
outcomes of the subset were assessed and reported indepen-
dently.
Studies on the effectiveness of analgesics given presurgery,
during surgery, or directly postsurgery and in which the effect
of these analgesics on pain as a result of the surgery was
studied are excluded from this review.
Study Selection
Two reviewers (M.S.R./S.G., B.M.H.) independently ap-
plied the inclusion criteria to select potential relevant studies
from the title and abstracts of the references retrieved by the
literature search. A consensus method was used to solve any
disagreements concerning inclusion of studies, and a third
reviewer (B.W.K.) was consulted if disagreement persisted.
RCTs published after the search data mentioned in the
(Cochrane) review and RCTs investigating interventions not
Arch Phys Med Rehabil Vol 91, July 2010
summarized in a (Cochrane) review were included in the
present study.
Categorization of the Relevant Literature
Relevant publications are categorized under 3 headers: Sys-
tematic reviews, Recent RCTs, and Additional RCTs. The
header “Systematic reviews” describes all Cochrane and
Cochrane-based systematic reviews. The header “Recent
RCTs” contains all RCTs published from the final date of the
search strategy that the systematic review covered. Finally, the
header “Additional RCTs” describes all RCTs concerning in-
terventions that have not yet been described in a systematic
review.
Data Extraction
Two researchers (M.S.R./S.G., B.M.H.) independently ex-
tracted the data. Information was collected on the study pop-
ulation, interventions used, outcome measures, and outcome. A
consensus procedure was used to solve any disagreement be-
tween the researchers.
The follow-up period was categorized into the short term
(0 –3mo), the midterm (4 – 6mo), and the long term (⬎6mo).
Methodologic Quality Assessment
To identify potential risks of bias of the included RCTs, 2
reviewers (M.S.R., B.M.H.) independently assessed the meth-
odologic quality of each RCT. The 12 quality criteria (table 1)
and operationalization of these criteria (appendix 2) were
adapted from Furlan et al.15 Each item was scored as “yes,”
“no,” or “don’t know.” High quality was defined as a score of
50% or more (ie, a “yes” score on 50% or more of the criteria)
on the methodologic quality assessment. A consensus proce-
dure was used to solve any disagreement between the review-
ers.
Data Synthesis
If quantitative analysis of the studies was not possible be-
cause of diverse outcome measures and other clinical hetero-
geneity, a meta-analysis was not performed. In that case, we
summarized the results using a rating system consisting of 5
levels of scientific evidence, taking into account the method-
ologic quality and the outcome of the original studies (best-
evidence synthesis).16 All RCTs together—that is, the number
of RCTs found in the reviews plus the number of recent RCTs
or the number of additional RCTs— determined the available
number of RCTs for a certain intervention. The article was
included in the best-evidence synthesis only if a comparison
was made between the groups (treatment vs placebo, treatment
vs control, or treatment vs another treatment) and the level of
significance was reported. The results of the study were labeled
“significant” if 1 of the 3 outcome measures had significant
results.
The level of evidence was ranked and divided into the
following levels:
1. Strong evidence for effectiveness: consistent (ⱖ75% of
the trials report consistent findings); positive (signifi-
cant) findings within multiple higher-quality RCTs
2. Moderate evidence for effectiveness: consistent positive
(significant) findings within multiple lower-quality
RCTs and/or 1 high-quality RCT
3. Limited evidence for effectiveness: positive (significant)
findings within 1 low-quality RCT
4. Conflicting evidence for effectiveness: provided by con-
flicting (significant) findings in the RCTs (⬍75% of the
trials report consistent findings)
 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede 983

Table 1: Methodologic Quality Criteria: Sources of Risk of Bias

Item Judgment

1. Was the method of randomization adequate? Yes / No / Unsure

2. Was the treatment allocation concealed? Yes / No / Unsure
Was knowledge of the allocated interventions adequately prevented during the study?
3. Was the patient blinded to the intervention? Yes / No / Unsure
4. Was the care provider blinded to the intervention? Yes / No / Unsure
5. Was the outcome assessor blinded to the intervention? Yes / No / Unsure
Were incomplete outcome data adequately addressed?
6. Was the dropout rate described and acceptable? Yes / No / Unsure
7. Were all randomized participants analyzed in the group to which they were allocated? Yes / No / Unsure
8. Are reports of the study free of suggestion of selective outcome reporting? Yes / No / Unsure
Other sources of potential bias:
9. Were the groups similar at baseline regarding the most important prognostic indicators? Yes / No / Unsure
10. Were co-interventions avoided or similar? Yes / No / Unsure
11. Was the compliance acceptable in all groups? Yes / No / Unsure
12. Was the timing of the outcome assessment similar in all groups? Yes / No / Unsure

5. No evidence found for effectiveness of the inventions: Effectiveness of Interventions

RCTs available, but no (significant) differences between
intervention and control groups were reported
6. No systematic review or RCT found
RESULTS
Characteristics of the Included Studies
The initial literature search resulted in the identification of 4
systematic reviews from the Cochrane Library and 47 reviews
(7 from PubMed, 29 from EMBASE, 11 from CINAHL). We
identified another 750 RCTs (241 from PubMed, 276 from
EMBASE, 177 from CINAHL, 56 from PEDro). Finally, after
selection based on the content of the titles, abstracts, and full
text of the references, 2 Cochrane reviews and 26 recent RCTs
(25 from PubMed 1 from PEDro, none from EMBASE or
CINAHL) met our inclusion criteria. No additional RCTs were
found. Four RCTs (2 from PubMed,17 2 from EMBASE18,19)
were initially included based on the content of their abstract.
Because the full texts were not available in national and inter-
national medical libraries, we contacted the authors by e-mail;
however, no full-text articles were received, so these articles
could not be included in the present review. The data extraction
of the included studies is presented in appendix 3 (systematic
reviews) and appendix 4 (recent RCTs).
Methodologic Quality of the Included Studies
The results of the methodologic quality assessment of the 20
included recent and additional RCTs are presented in table 2.
The Cochrane review of O’Connor et al13 (which reported on
nonsurgical treatment other than steroid injections) used the
methodologic quality criteria of the Cochrane Reviewers Hand-
book 4.0.20 Eight quality items were described, and RCTs were
defined as high-quality (A), moderate-quality (B), or low-
quality (C). Moderate RCTs had a score of 50% or more on the
quality criteria. Therefore, we decided that A and B study
scores would be defined as high-quality RCTs (table 3). The
methodologic quality criteria of Jadad et al21 were used in the
Cochrane review of Marshall et al14 reporting on corticosteroid
injections. Five quality items were described, and they defined
poor-quality and good-quality studies (table 4).
A total of 53 RCTs are included in our systematic review. Of
these, 29 RCTs (55%) were of high quality, and 8% of the
studies scored 40% to 50% of the total score.
Strong and moderate evidence for the effectiveness of non-
surgical interventions for the treatment of CTS is presented in
table 5. A complete overview of levels of evidence for the
effectiveness of all the identified nonsurgical interventions is
presented in table 6.
1. Nonsurgical Treatment (other than steroid injections)
The Cochrane review of O’Connor13 (search up to February
2001, PubMed; up to March 2002, EMBASE; up to December
2001, CINAHL and PEDro) included 21 trials (n⫽923) study-
ing the effectiveness of all types of nonsurgical treatments
(other than steroid injections) for CTS. The trials presented
findings in 12 treatment areas: splinting, ultrasound, ergonomic
keyboards, oral medication, vitamins, exercise, yoga, mobili-
zation, magnet therapy, chiropractic care, laser, and acupunc-
ture. Furthermore, we found 18 recent RCTs (n⫽963) on the
effectiveness of splinting, ultrasound, laser, oral medication,
manual therapy, magnetic field stimulation, acupuncture, mas-
sage therapy, heat wrap therapy, cupping therapy, botulinum B
toxin, iontophoresis, and exercise. No additional RCTs were
found.
1.1. Splinting
In the systematic review of O’Connor,13 3 RCTs22-24 on
splinting were included. Furthermore, 7 recent RCTs25-31 on
splinting were found.
Different Positions for a Wrist Splint Compared
Systematic review. One low-quality RCT22 (n⫽90) in the
Cochrane review of O’Connor13 compared the short-term ef-
fects of a wrist splint in neutral position with a wrist splinted in
20° extension. After 2 weeks of treatment, significant overall
and nocturnal improvement (RR⫽2.43, 95% CI, 1.12–5.28;
and RR⫽2.14, 95% CI, .99 – 4.65, respectively) was found in
favor of the splint in neutral position.
We concluded that there is limited evidence that the use of
a wrist splint in neutral position is more effective than an
extended wrist position of 20° in patients with CTS in the short
term (2 weeks).
Nocturnal Hand Brace Versus No Treatment
Systematic review. O’Connor13 found 1 low-quality RCT23
(n⫽80) that compared a nocturnal hand brace with no treatment.

Arch Phys Med Rehabil Vol 91, July 2010

Arch Phys Med Rehabil Vol 91, July 2010

984
Table 2: Methodologic Quality Scores of the Included RCTs
Incomplete Free of Timing of
Outcome Incomplete Suggestions the
Blinding? Data Outcome of Selective Similarity of Co-interventions Compliance Outcome
Adequate Allocation Blinding? Blinding? Outcome Addressed? Data? ITT Outcome Baseline Avoided or Acceptable in Assessment Score Study
Reference Randomization? Concealment? Patients? Caregiver? Assessors? Dropouts? Analysis? Reporting? Characteristics? Similar? All Groups? Similar? Maximum Score Percentage

EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede

Dammers
et al73 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ? NA ⫹ 11 10 90
Chang et al48 ? ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ? ⫹ 12 10 83
Irvine et al39 ⫹ ⫹ ⫹ ⫹ ? ⫹ ⫹ ⫹ ⫹ ⫹ ? ⫹ 12 10 83
Bialosky
et al97 ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫺ ⫹ ⫹ ? ⫹ ⫹ 12 9 75
Amirjani
et al64 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ? ? ? ⫹ 12 8 67
Bakhtiary and
Rashidy-
Pour38 ⫹ ⫹ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ? ? ⫹ 12 8 67
Brininger
et al26 ⫹ ⫹ ⫺ ⫺ ? ⫹ ⫹ ⫹ ⫹ ? ⫹ ⫹ 12 8 67
Evcik et al40 ⫹ ? ⫹ ? ⫹ ⫹ ⫹ ⫹ ? ⫹ ? ⫹ 12 8 67
Hui et al69 ⫹ ⫹ ⫹ ⫹ ⫹ ? ? ⫹ – ⫹ ? ⫹ 12 8 67
Mishra et al33 ⫹ ? ⫺ ⫺ ? ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ 12 8 67
Burke et al50 ⫹ ⫹ ⫺ ⫺ ⫹ ⫹ ? ⫹ ? ? ⫹ ⫹ 12 7 58
Yagci et al30 ⫹ ⫹ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ⫹ ? ? ⫹ 12 7 58
Yang et al57 ⫹ ⫹ ⫺ ⫺ ? ⫹ ⫹ ⫹ ⫹ ? ? ⫹ 12 7 58
De Angelis
et al31 ? ⫺ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ? ⫹ ⫹ ⫹ 12 6 50
Michalsen
et al61 ⫹ ⫺ ⫺ ⫺ ? ⫹ ⫹ ⫹ ? ⫹ ? ⫹ 12 6 50
Shooshtari
et al41 ? ? ⫹ ? ? ⫹ ⫹ ⫹ ? ⫹ ? ⫹ 12 6 50
Weintraub
and Cole55 ⫹ ? ⫹ ? ? ⫹ ⫺ ⫹ ? ⫹ ? ⫹ 12 6 50
Baysal et al25 ⫹ ⫹ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ? ? ⫺ 12 5 42
Moghtaderi
et al79 ? ? ⫺ ⫺ ? ⫹ ? ⫹ ⫹ ? ? ⫹ 12 5 42
Michlovitz
et al60 ? ? ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ – ? ? ⫹ 12 5 42
Moraska
et al58 ? ? ⫹ ⫺ ⫺ ⫹ ⫹ ⫹ – ? ? ⫹ 12 5 42
Premoselli
et al29 ⫺ ? ⫺ ? ⫹ ⫹ ⫺ ⫹ – ? ⫹ ⫹ 12 5 42
Breuer et al62 ? ? ⫺ ? ? ⫹ ⫹ ⫹ ⫺ ? NA ⫹ 11 4 36
Pinar et al28 ? ? ⫺ ⫺ ? ⫹ ⫹ ⫹ ? ? ? ⫹ 12 4 33
Heebner and
Roddey27 ⫹ ? ⫺ ⫺ ? ⫺ ⫺ ⫹ ? ? ? ⫹ 12 3 25
Field et al59 ? ? ⫺ ⫺ ? ? ? ⫹ ? ? ? ⫹ 12 2 17

Abbreviations: ⫹, Yes; ⫺, No; ?, unsure; ITT, intention to treat; NA, not applicable (in an intervention such as surgery, compliance is not an issue).
 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede 985

Table 3: Methodologic Quality Scores of the Cochrane Review of O’Connor et al13

Quality Our
Score Definition of
Blinding No No No According to High or Low
Allocation Blinding Blinding Outcome Selection Performance Attrition O’Connor Score Study Quality of
13
Reference Randomization? Concealment? Patients? Caregiver? Assessors? Bias? Bias? Bias? et al Maximum Score Study

Ebenbichler
et al35 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ A 8 8 High
Hui et al44 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ A 8 8 High
Spooner
et al46 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ A 8 8 High
Carter
et al54 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫾ B 8 7 High
Chang
et al42 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫾ ⫹ B 8 7 High
Herskovitz
et al43 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫾ ⫹ B 8 7 High
Rempel
et al52 ⫹ ⫹ ⫹ ⫹ ⫹ ⫾ ⫹ ⫾ B 8 6 High
Aigner
et al56 ⫹ ⫹ ⫹ ⫺ ⫹ ⫾ ⫹ ⫹ B 8 5 High
Ozkul
et al63 ⫹ ? ⫹ ⫹ ⫺ ⫾ ⫹ ⫹ B 8 5 High
Oztas
et al36 ⫹ ? ⫹ ⫺ ⫺ ⫾ ⫾ ⫹ B 8 3 High
Pal et al45 ⫹ ? ⫹ ⫺ ⫹ ⫾ ⫹ ⫾ B 8 4 High
Davis
et al51 ⫹ ⫹ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ C 8 5 Low
Manente
et al23 ⫹ ⫹ ⫺ ⫺ ⫺ ⫹ ⫺ ⫹ C 8 4 Low
Burke
et al22 ⫹ ? ⫹ ⫺ ⫹ ⫺ ⫺ ⫾ C 8 3 Low
Garfinkel
et al34 ⫹ ⫹ ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ C 8 3 Low
Koyuncu
et al37 ⫹ ? ⫹ ⫺ ⫹ ⫾ ⫾ ⫹ C 8 3 Low
Stransky
et al47 ⫹ ? ⫹ ⫺ ⫹ ⫾ ⫺ ⫺ C 8 3 Low
Tal-Akabi
and
Rushton49 ⫹ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ C 8 3 Low
Akalin
et al32 ⫹ ? ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ C 8 2 Low
Tittiranonda
et al53 ⫹ ? ⫺ ⫺ ⫹ ⫾ ⫾ ⫺ C 8 2 Low
Walker
et al24 ⫹ ? ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ C 8 2 Low
13
NOTE. Definition of O’Connor et al: A, high quality: all criteria met; B, moderate quality: 1 or more criteria partly met; C, low quality: 1 or more
criteria not met.
Abbreviations: ⫹, Yes; ⫺, No; ?, unsure; ⫾, partly met.

Significant results were found in favor of a nocturnal hand brace
compared with no treatment on symptom improvement
(WMD⫽⫺1.07; 95% CI, ⫺1.29 to ⫺.85), hand function
(WMD⫽⫺.55; 95% CI, ⫺.82 to ⫺.28), and overall improvement
(RR⫽4.00; 95% CI, 2.34 – 6.84) at 4 weeks of follow-up.
Recent RCTs. In the low-quality RCT of Premoselli et
al,29 (n⫽50) the positive results found in the review of
O’Connor13 were confirmed at 3 and 6 months of follow-up:
significantly better results in favor of a nocturnal neutral wrist
splint were found on symptoms (mean differences ⫾ SD,
splint, 1.07⫾.39, vs control, ⫺.02⫾.24, P⫽.001; and splint,
1.22⫾.39, vs control, .17⫾.29, P⫽.001, respectively) and
function (mean differences ⫾ SD, splint, .53⫾.22, vs control,
⫺.15⫾.43, P⫽.0001; and splint, .75⫾.28, vs control, .04⫾.30,
P⫽.0004, respectively).
It was concluded that there is moderate evidence in the short
term and limited evidence in the midterm that a nocturnal hand
brace is more effective than no therapy in the treatment of
patients with CTS.
Full-time use of a Wrist Splint Versus Night-Only Use
Systematic review. In the low-quality RCT of Walker et
al24 (n⫽24) included in the review of O’Connor13 that com-
pared the full-time use of a wrist splint with night-only use, no
significant differences were found on symptom improvement

Arch Phys Med Rehabil Vol 91, July 2010

986 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede

Table 4: Methodologic Quality Scores of the Cochrane Review of Marshall et al14

Allocation Withdrawal/ Method Method Score Study Quality of
Reference Randomization? Concealment? Double-Blind? Dropouts? Randomizing? Blinding? Maximum Score Percentage Study*

Armstrong
et al66 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ 6 6 100 Good
Dammers
et al65 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ 6 6 100 Good
Wong et al68 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ 6 6 100 Good
Celiker et al70 ⫹ ⫹ ⫺ ⫹ ⫹ ⫺ 6 4 67 Good
Wong et al74 ⫹ ? ⫹ ⫹ ⫺ ⫹ 6 4 67 Good
Ozdogan and
Yazici67 ⫹ ? ⫹ ⫺ ⫺ ⫹ 6 3 50 Good
Aygul et al77 ⫹ ? ⫺ ⫹ ⫺ ⫺ 6 2 33 Poor
Habib et al76 ⫹ ? ⫺ ⫹ ⫺ ⫺ 6 2 33 Poor
Sevim et al75 ⫹ ? ⫺ ⫺ ⫺ ⫹ 6 2 33 Poor
Gökoğlu et al78 ⫹ ? ⫺ ⫺ ⫺ ⫺ 6 1 17 Poor
Lucantoni
et al71 ⫹ ? ⫺ ⫺ ⫺ ⫺ 6 1 17 Poor
O’Gradaigh and
Merry72 ⫹ ? ⫺ ⫺ ⫺ ⫺ 6 1 17 Poor

Abbreviations: ⫹, Yes; ⫺, No; ?, unsure.

*Good, high quality; Poor, low quality.

(WMD⫽⫺.21; 95% CI, ⫺.83 to .41) or hand function Wrist Splint Versus Hand Brace
(WMD⫽⫺.21; 95% CI, ⫺.87 to .45) at 6 weeks of follow-up. Recent RCTs. One recent high-quality RCT (n⫽120)31
In conclusion, there is no evidence for the effectiveness of a compared a wrist splint with a hand brace. Both groups wore
full-time use of a wrist splint compared with night-only use in the orthotic devices for 3 months at night. No significant
patients with CTS in the short term.
differences between the groups were found on the symptom
severity score, on the function severity score of the Boston
Carpal Tunnel Questionnaire, and on pain from baseline to 3
Table 5: Strong and Moderate Evidence for Effectiveness months of follow-up and to 9 months of follow-up.
of Nonsurgical Interventions for CTS
Therefore, there is no evidence for the effectiveness of a
Nonsurgical Interventions to Treat Strong or Moderate night hand brace compared with night splinting of the wrist for
CTS Evidence Found the treatment of CTS in the short term.
✓abc
Physiotherapy
✓de
Oral Tendon and Nerve Gliding Exercises as Additive to
✓fghij
Injection Splinting
✓klmno
Other nonsurgical interventions
Systematic review. One low-quality trial32 (n⫽36) found
✓
Strong or moderate evidence found. no significant differences on symptom improvement, hand
function, grip strength, and pinch strength for nerve and tendon
Short-term: gliding exercises as additive to a neutral wrist splint at 3
a
Moderate evidence: ultrasound* vs placebo at 7wk of follow-up.
c
Moderate evidence: ultrasound* vs laser.
months of follow-up.
d
Strong evidence: oral steroids* vs placebo at 2wk of follow-up. Recent RCTs. The low-quality study of Baysal et al25
e
Moderate evidence: oral steroids* vs placebo at 4wk of follow-up. (n⫽56) reported on ultrasound, splinting, and nerve and tendon
f
Strong evidence: corticosteroid injections* vs placebo. gliding exercises. Patients were divided into 3 treatment
g
Moderate evidence: local* vs systemic steroids injection.
h
Moderate evidence: local corticosteroid injection* vs oral steroids.
groups. Group 1 was treated with a splint and nerve and tendon
i
Moderate evidence: insulin injections as additive to steroids gliding exercises, group 2 with a splint and ultrasound treat-
injections in patients with noninsulin-dependent diabetes ment, and group 3 with a splint, nerve and tendon gliding
mellitus. exercises, and ultrasound treatment. No results between the
k
Moderate evidence: nocturnal hand brace* vs no therapy.
l
Moderate evidence: wrist splinting vs prednisone.*
groups were presented. However, within the 3 treatment
m
Moderate evidence: ergonomic keyboard* vs standard keyboard. groups, significant differences were found on pain (VAS), grip
n
Moderate evidence: dynamic magnetic field therapy* vs placebo strength, pinch strength, the symptom severity scale, and the
therapy. function status scale at 8 weeks of follow-up.
o
Moderate evidence: cupping therapy vs head† pads.
One recent high-quality RCT26 (n⫽51) studied 2 types of
Midterm:
b
splints (a neutral wrist and metacarpophalangeal splint, group
Moderate evidence: ultrasound*⬎ vs placebo. 1; and a wrist cock-up splint, group 2) with and without tendon
j
Moderate evidence: 60mg methylprednisone* vs 20 or 40mg
methylprednisone in the midterm. and nerve gliding exercises (groups 3 and 4, respectively).
Significant results within the groups (no P value given) on
Long-term: symptoms (group 1, 38%, vs group 2, 17%; groups 3 and 4, no
percentages given) at 8 weeks of follow-up and pinch strength
– (no further data given) within the groups were found at 4 weeks
*in favor of follow-up. No between-group results were given.

Arch Phys Med Rehabil Vol 91, July 2010

 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede 987

Table 6: Total Overview of Evidence for Effectiveness of Nonsurgical Interventions for CTS
Nonsurgical Treatment

Physiotherapy Oral Treatment Injection Other Nonsurgical Treatments

Ultrasound –Steroids vs Corticosteroid injections Splinting

–Ultrasound* vs Placebo –Corticosteroid injection vs –Splinting in neutral position* vs splinting in extended wrist
placebo Short-term: placebo position of 20°
Short-term: 2wk: ⫹⫹⫹ Short-term: ⫹⫹⫹ Short term (2wk): ⫹
2wk: NE 4wk: ⫹⫹ –Local* vs systemic –Nocturnal hand brace* vs no therapy
7wk: ⫹⫹ –Nonsteroidal Short-term: ⫹⫹ Short-term: ⫹⫹
Midterm: ⫹⫹ anti- –Corticosteroid injection* Midterm: ⫹
–1.5W/cm2 vs inflammatory vs oral steroids: –Full-time use wrist splint vs night-only splint
0.8W/cm2 drugs vs Short-term: ⫹⫹ Short-term: NE
Short-term: NE placebo Long-term: NE –Night wrist splint vs night hand brace
–1 vs 3MHz Short-term: NE –Corticosteroid injection vs Short-term: NE
Short-term: NE –Diuretica vs anti-inflammatory –6wk day-and-night splint followed by 4wk night-splint and
–Ultrasound* vs placebo medication plus splint 4wk nerve gliding exercises* vs same treatment without
laser Short-term: NE Short-term: NE nerve gliding exercises
Short-term: ⫹⫹ –Vitamin B6 –Corticosteroid injection* Short-term: ⫹
vs placebo vs Helium laser treatment –Active neurodynamic exercises* as additive to night splint
Laser therapy Short-term: NE Short-term: ⫹ during heavy activities plus tendon gliding exercises
–Laser vs –Oral Midterm: NE Midterm: ⫹
placebo Prednisone –25mg hydrocortisone vs –Neutral wrist plus MCP splint (NW) vs wrist cock-up splint
Short-term: NE 4 wk vs oral 100mg hydrocortisone (WCP) vs NW plus tendon and nerve gliding exercises (E)
prednisone Short-term: NE vs WCP plus E
Mobilization and 2wk: –60mg* Methylprednisone Short-term: NE
manual therapy Long-term: NE vs 20 or 40mg –Splint plus nerve and tendon gliding exercises (NTE) vs
–Carpal bone Methylprednisone splint plus ultrasound vs splint plus NTE plus ultrasound
mobilization* Midterm: ⫹⫹ Short-term: NE
vs no treatment Long-term: NE –Wrist splint vs oral prednisone*
Short-term: ⫹ –Short vs long-acting Short-term: ⫹⫹
–Neurodynamic corticosteroid injection –Low-level laser as additive to splinting
vs carpal bone Short-term: NE Short-term: NE
mobilization –Single vs 2 local –Yoga vs wrist splinting
Short-term: NE corticosteroid injections Short-term: NE
–Neurodynamic (15mg methylprednisone)
technique plus Short-term: NE Chiropractic treatment
splinting vs Midterm: NE –Chiropractic treatment vs medical treatment
sham therapy Long-term: NE Midterm: NE
plus splinting –Novel approach vs classic
Short-term: NE approach of injection Ergonomic keyboards
–Graston Short-term: NE –Ergonomic keyboard* vsstandard keyboard
instrument– –Proximal approach vs Short-term: ⫹⫹
assisted soft distal approach of –Apple keyboard* vs standard keyboard
tissue injection Midterm: ⫹
mobilization Long-term: NE –Microsoft keyboard* vs standard keyboard
(GISTM) plus –Corticosteroid injection* Midterm: ⫹
home exercises vs iontophoresis –Other ergonomic keyboards vs regular keyboard
vs manual soft Short-term: ⫹ Midterm: NE
tissue –Corticosteroid injection vs
mobilization by phonophoresis Magnet therapy
a clinician plus Midterm: NE –Magnet therapy vs placebo
home exercises –Corticosteroid injection vs Short-term: NE
Midterm: NE EMLA cream
Short-term: NE Magnetic field therapy
Massage –Dynamic magnet field therapy vs placebo
–Targeted Injections other than Short-term: ⫹⫹
massage steroid
protocol* vs –Botulinum B toxin vs Acupuncture
general placebo –Laser acupuncture vs placebo
massage Midterm: NE Short-term: NE
protocol –Acupuncture vs oral steroids
Short-term: ⫹ Insulin as additive to Short-term: NE
–Massage steroid injection
therapy for 15 –In noninsulin-dependent Heat wrap therapy
min plus self- diabetes mellitus: steroid –Heat wrap therapy* vs oral placebo
massage vs no injections followed by Short-term (3d): ⫹
treatment NPH injection* vs steroid
Short-term: ⫹ injections followed by Cupping therapy
placebo injections –Cupping therapy* vs heat pads
Short-term: ⫹⫹ Short-term (7d): ⫹⫹

Iontophoresis
–Dexamethasone iontophoresis vs Placebo
Midterm: NE
Long-term: NE

Abbreviations: ⫹, limited evidence found; ⫹⫹, moderate evidence found; ⫹⫹⫹, strong evidence found; d, days; EMLA, Eutectic mixture of Local Anesthetic; MCP, metacarpo-
phalangeal; mo, month; NE, no evidence found for effectiveness of the treatment: RCTs available, but no differences between intervention and control groups were found; NPH,
isophane insulin injection; vs, versus; wk, weeks.
*In favor of.

Arch Phys Med Rehabil Vol 91, July 2010

988 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
The low-quality RCT of Pinar et al28 (n⫽35) compared 2
groups of patients with CTS: a day-and-night splint and a
nonsurgical training program (nerve and gliding exercises)
were applied for 6 weeks to both groups. Subsequently, a night
splint only was used in both groups, and nerve and gliding
exercises were continued in the experimental group for the
remaining 4 weeks. Significant progress was detected on grip
strength between the experimental group and the control group
(mean ⫾ SD, 4.2⫾4.1 vs 1.3⫾1.5, respectively) at 10 weeks of
follow-up in favor of the experimental group. Furthermore,
between-group analyses showed no significant differences on
pain and pinch strength.
The low-quality study of Heebner and Roddey27 (n⫽60)
investigated active neurodynamic exercises as additive to stan-
dard care (consisting of splinting at night during heavy activ-
ities plus tendon gliding exercises). No significant differences
on the DASH Questionnaire, symptom severity score, and
neurodynamic irritability of the median nerve were found at 6
months of follow-up. Significant differences were found on the
function severity scale in favor of standard care with active
neurodynamic exercises at 6 months of follow-up (standard
care with active neurodynamic exercises, 2.2 [mean], com-
pared with standard care, 2.9; P⫽.016).
In conclusion, there is limited evidence that 6 weeks of
day-and-night splinting with a nonsurgical training program
followed by 4 weeks of night splint with nerve gliding exer-
cises is more effective than the same treatment without nerve
gliding exercises in the short term, and that active neurody-
namic exercises as additive to standard care (ie, night splint
during heavy activities plus tendon gliding exercises) is more
effective (limited evidence) than standard care alone in the
midterm. There is no evidence for the effectiveness of a neutral
wrist and metacarpophalangeal splint compared with a wrist
cock-up splint with and without tendon and nerve gliding
exercises. Furthermore, there is no evidence for the effective-
ness of treatment with a splint plus nerve and tendon gliding
exercises compared with treatment with a splint plus ultrasound
or compared with treatment with a splint plus tendon gliding
exercises plus ultrasound in the short term.
Splinting Versus Oral Prednisone
Recent RCTs. One recent high-quality RCT33 (n⫽71)
compared splinting of the wrist in neutral position for 4 weeks
with oral prednisolone 20mg/d for 2 weeks followed by
10mg/d for 2 weeks. Significant differences were reported on
the function status score in favor of oral steroids compared with
splinting (mean ⫾ SD, splint, .16⫾.17, vs oral steroids,
.26⫾.21; P⫽.03), but no significant differences were found on
the symptom severity scale at 3 months of follow-up.
In conclusion, there is moderate evidence that oral steroids
are more effective than splinting of the wrist to treat CTS in the
short term.
Splinting Versus Splinting Plus Low-Level Laser Therapy
Recent RCTs. One recent high-quality RCT30 compared a
full-time hand splint in neutral position for 3 months with
splinting plus 10 sessions of low-level laser therapy. Only
significant within-group results were reported on the symptom
severity score of the Boston Carpal Tunnel Questionnaire and
for grip strength within the splinting group. No significant
differences between the groups were found on the Boston
Carpal Tunnel Questionnaire (function capacity and symptom
severity scores). At 3 months of follow-up, no comparison
between the groups for grip strength was made.
Arch Phys Med Rehabil Vol 91, July 2010
Therefore, there is no evidence for the effectiveness of
splinting compared with splinting plus low-level laser therapy
in the short term.
Splinting Versus Yoga
Systematic review. One low-quality RCT (n⫽51)34 in the
review of O’Connor13 compared yoga with wrist splinting and
found no significant differences on pain at 8 weeks of
follow-up.
In conclusion, there is no evidence for the effectiveness of
yoga compared with wrist splinting to treat CTS in the short
term.
1.2 Ultrasound
Ultrasound Versus Placebo
Systematic review. An analysis of pooled data from 2
trials35,36 (n⫽63) of ultrasound treatments compared with pla-
cebo of O’Connor13 showed no significant effects on pain,
symptoms, or function at 2 weeks of follow-up. However, 1
high-quality trial35 showed significant symptom improvement
after 7 weeks in patients treated with ultrasound
(WMD⫽⫺.99; 95% CI, ⫺1.77 to ⫺.21), which was main-
tained at 6 months of follow-up (WMD⫽⫺1.86; 95% CI,
⫺2.67 to ⫺1.05).
Thus, there is no evidence for the effectiveness of ultrasound
compared to placebo at 2 weeks of follow-up, but there is
moderate evidence that ultrasound is more effective than pla-
cebo in the treatment of patients with CTS at 7 weeks of
follow-up and in the midterm.
Ultrasound: Comparison of Intensities
Systematic review. One high-quality RCT36 (n⫽30) in-
cluded in the review of O’Connor13 compared 2 intensities of
ultrasound (1.5W/cm2 and 0.8W/cm2) but found no significant
differences regarding pain and symptom improvement between
these intensities after 2 weeks.
Therefore, we conclude there is no evidence for the effec-
tiveness of an ultrasound intensity of 1.5W/cm2 compared with
0.8W/cm2 in the short term.
Ultrasound: Different Frequencies Compared
Systematic review. At 4 weeks of follow-up in another
low-quality RCT37 (n⫽21) included in the review of
O’Connor,13 2 different frequencies (1 and 3MHz) were com-
pared, but no significant differences were found on pain and
function. It was concluded that there is no evidence for the
effectiveness of 1 or 3MHz frequency of ultrasound in patients
with CTS in the short term.
Ultrasound Versus Laser Therapy
Recent RCTs. In a high-quality RCT38 (n⫽90), ultra-
sound was compared with laser therapy. Ultrasound ap-
peared to be significantly more effective than laser therapy
on pain (MD between groups, ⫺4.4; 95% CI, ⫺4.9 to ⫺3.1;
P⬍.001) and function (hand grip strength: MD between
groups, 12.1; 95% CI, 5.7–27.6; P⬍.001) at 4 weeks of
follow-up.
Therefore, there is moderate evidence that ultrasound is
more effective than laser therapy in the treatment of patients
with CTS in the short term.
1.3 Laser Therapy
Recent RCTs. Three recent RCTs on laser therapy were
found. In the high-quality RCT of Irvine et al39 (n⫽15), no
significant differences were found between low-level laser
 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
therapy and placebo on a symptom severity scale and on hand
performance score at 9 weeks of follow-up.
In the high-quality study of Evcik et al40 (n⫽81), no signif-
icant differences were found between the low-level laser ther-
apy and placebo treatment on hand grip strength, pinch grip,
pain, and functional capacity at 12 weeks of follow-up.
The high-quality study of Shooshtari et al41 (n⫽80) com-
pared low-level laser therapy to placebo. No comparisons be-
tween the 2 groups were made. Significant differences were
found on pain and hand grip within the low-level laser therapy
group (mean ⫾ SD, from 7.8⫾.42 before treatment to
4.98⫾.12 at 3 weeks of follow-up, P⬍.001; and from
19.81⫾5.06kg before treatment to 22.86⫾5.13kg at 3 weeks of
follow-up, P⬍.001, respectively). Within the placebo group,
significant differences were found for pain (mean ⫾ SD, from
8.01⫾.36 before treatment to 7.62⫾0.4 at 3 weeks of follow-
up; P⬍.001), but no significant differences were found for
hand grip.
In conclusion, there is no evidence for the effectiveness of
laser therapy compared with placebo as an intervention to treat
CTS in the short term.
1.4 Oral Medications and Vitamins
Systematic review. Six RCTs42-47 (n⫽243) in the Co-
chrane review of O’Connor13 reported on oral medication or
vitamins. Three high-quality RCTs compared oral steroids with
placebo. 42-44 Pooling of the data of these 3 trials demonstrated
significant changes in favor of oral steroids on symptom im-
provement (WMD⫽⫺7.23; 95% CI, ⫺10.31 to ⫺4.14) at 2
weeks of follow-up. One RCT42 found significant differences
on symptom improvement at 4 weeks of follow-up
(WMD⫽⫺10.8; 95% CI, ⫺15.26 to ⫺6.34).
Two other high-quality RCTs compared nonsteroidal anti-
inflammatory drugs42 and diuretics45 with placebo. No signif-
icant benefit on symptom improvement was reported for non-
steroidal anti-inflammatory drugs or diuretics versus placebo at
4 weeks of follow-up. One high-quality study46 and 1 low-
quality study 47 found no significant differences between vita-
min B6 and placebo on overall symptoms at 10 to 12 weeks of
follow-up.
Recent RCTs. The long-term effects of the study of Chang
et al42 included in the Cochrane review of O’Connor13 were
reported by the high-quality RCT of Chang et al48 (n⫽109).
Chang48 compared oral prednisolone given for 4 weeks (20mg
daily for 2 weeks followed by 10mg daily for 2 weeks) with
oral prednisolone given for 2 weeks (20mg daily for 2 weeks
and placebo for 2 weeks). No significant differences on overall
improvement were found at 12 months of follow-up.
In conclusion, there is strong evidence after 2 weeks and
moderate evidence after 4 weeks that oral steroids are more
effective than placebo. There is no evidence for the effective-
ness of 20mg daily of prednisolone for 2 weeks followed by
10mg daily of the same drug for 2 weeks versus 20mg pred-
nisolone daily for 2 weeks followed by placebo in the long
term. Furthermore, there is no evidence for the effectiveness of
anti-inflammatory drugs or diuretica in the short term. In ad-
dition, there is no evidence for the effectiveness of vitamin B6
to treat CTS in the short term.
1.5 Other Nonsurgical Treatments
Mobilization and Manual Therapy
Systematic review. The low-quality RCT of Tal-Akabi and
Rushton49 (n⫽21) on carpal bone mobilization demonstrated a
significant benefit on symptoms compared with no treatment
(WMD⫽⫺1.43; 95% CI, ⫺2.19 to ⫺.67) at 3 weeks of follow-
989
up. Further, no significant results were found on any outcome
regarding pain, function, or improvement comparing neurody-
namic with carpal bone mobilization after 3 weeks of follow-
up. No significant results were found on any outcome regarding
function by comparing neurodynamic with carpal bone mobi-
lization in the short term.
Recent RCTs. The high-quality study of Bialosky et al97
reported on a neurodynamic technique intended to provide
anatomic stress across the median nerve and combined this
intervention with splinting for 3 weeks. This intervention was
compared with sham therapy. Both groups were also treated
with a splint for 3 weeks. After 3 weeks of treatment, no
significant differences between the groups were found on pain,
the DASH Questionnaire, or grip strength.
The high-quality study of Burke et al50 (n⫽22) compared 2
manual therapy interventions: the Graston instrument–assisted
soft tissue mobilization plus home exercises with manual soft
tissue mobilization by a clinician plus home exercises. Im-
proved results were found within groups, but there were no
significant differences between the 2 groups on pain, range of
motion (flexion and extension), grip strength, and the Boston
Carpal Tunnel Questionnaire (functional status scale and the
symptom severity scale) at 6 months of follow-up.
Therefore, there is limited evidence that carpal bone mobi-
lization is more effective than no treatment in the short term.
No evidence was found for the effectiveness of neurodynamic
versus carpal bone mobilization in the short term, for the
effectiveness of a neurodynamic technique plus splinting com-
pared with a sham therapy plus splinting group in the short
term, or for the effectiveness of Graston instrument–assisted
soft tissue mobilization plus home exercises compared with
soft tissue mobilization plus home exercises to treat CTS in the
midterm.
Chiropractic Treatment
Systematic review. No significant differences on hand
function between chiropractic treatment (ie, manual thrusts,
myofascial massage and loading, ultrasound, and nocturnal
wrist splint) and medical treatment (ie, ibuprofen and wrist
splint) were found in a low-quality trial of Davis et al51 (n⫽91)
at 13 weeks of follow-up.
Therefore, there is no evidence for the effectiveness of
chiropractic therapy compared with medical treatment for CTS
in the midterm.
Ergonomic Keyboards
Systematic review. Two RCTs52,53 included in the review
of O’Connor13 studied ergonomic keyboards compared with
control. The high-quality study of Rempel et al52 (n⫽18)
compared an ergonomic keyboard with a standard keyboard
and found significant changes on pain and hand function in
favor of the ergonomic keyboard (WMD⫽⫺2.40, 95% CI,
⫺4.45 to ⫺0.35; WMD⫽⫺2.20, 95% CI, ⫺12.08 to 7.68,
respectively) at 3 months of follow-up. The low-quality study
of Tittiranonda et al53 (n⫽80) found no significant differences
on pain among 3 ergonomic keyboards (ie, comfort keyboard
system, Apple adjusTable keyboard, and Microsoft natural
keyboard) and a regular keyboard at 6 months of follow-up. At
6 months of follow-up, significant changes on hand function
were found in favor of the Apple keyboard and the Microsoft
keyboard compared with a regular keyboard (WMD⫽.93, 95%
CI, .26 –1.60; WMD⫽1.92, 95% CI, .84 –3.00, respectively),
but no significant differences were found on hand function in
the ergonomic keyboard group.
Thus, there is moderate evidence that an ergonomic key-
board is more effective than a standard keyboard in the short
Arch Phys Med Rehabil Vol 91, July 2010
990 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
term. In the midterm, there is limited evidence that an Apple
keyboard and a Microsoft keyboard are more effective than a
regular keyboard, but no evidence for the effectiveness of other
ergonomic keyboards compared with a regular keyboard.
Magnet Therapy
Systematic review. One high-quality RCT (n⫽30)54 com-
pared magnet therapy with placebo and found no significant
benefit on pain between these groups at 2 weeks of follow-up.
Therefore, we found no evidence for the effectiveness of
magnet therapy.
Magnetic Field Therapy
Recent RCTs. Significant differences were found in a
high-quality study of Weintraub and Cole55 (n⫽36) on the
Neuropathy Pain Scale (total composite; reduction: treatment
group, 42%, compared with controls, 24%; P⫽.04) between
simultaneous and time-varying dynamic magnetic field stimu-
lation on the wrist and sham therapy from baseline to 2 months
of follow-up. In contrast, no significant differences were found
on pain and Patients Clinical Global Impression of Change at
2 months of follow-up.
Therefore, we found moderate evidence for the effectiveness
of dynamic magnetic field therapy in the short term to treat
patients with CTS.
Acupuncture
Systematic review. A high-quality RCT (n⫽26)56 demon-
strated no significant differences between laser acupuncture
and placebo on night pain at 3 weeks of follow-up.
Recent RCTs. The high-quality study of Yang et al57 com-
pared 4 weeks of acupuncture (8 sessions) with oral steroids
(first 2 weeks, 20mg prednisolone daily, followed by 2 weeks
of 10mg prednisolone daily). Both interventions resulted in
better but no significant differences on the Global Symptom
Score at 4 weeks of follow-up (mean percent change ⫾ SD
from baseline to 4 weeks, acupuncture group, ⫺70⫾24.6, vs
steroid group, ⫺64.7⫾27.6).
It was concluded that there is no evidence for the effective-
ness of laser acupuncture for the treatment of CTS in the short
term, or for the effectiveness of acupuncture compared with
oral steroid drugs to treat CTS in the short term.
Massage Therapy
Recent RCTs. The low-quality study RCT of Moraska et
al58 (n⫽27) compared a targeted massage protocol (focused on
the affected upper extremity and addressing areas of constric-
tion, ischemia, and nerve entrapment) with a general massage
protocol (relaxing massage to reduce tension of the back, neck,
and upper extremities) for 6 weeks. Significant effects were
found on grip strength at 10 weeks of follow-up in favor of the
targeted massage group (targeted massage group, mean from
25.1kg to 29.5kg; 95% CI, 27.7–31.3kg; vs the general mas-
sage group, mean from 25.1kg to 26.3kg; P⫽.04). No signif-
icant differences were found on pinch strength (at 6wk), symp-
tom severity score (at 10wk), function status scale (at 6wk),
and the Grooved Pegboard Test (at 6wk). The low-quality
study of Field et al59 (n⫽16) also examined massage therapy as
treatment for CTS, but they compared a 15-minute massage
once a week for a 4-week period plus self-massage daily with
a control group without treatment. Significant differences were
found in favor of the massage group on CTS (ie, loss of
strength, tingling, numbness, burning, or pain to the affected
region; massage group, from 3.00 at the first day to 2.22 at the
last day compared with controls, from 3.00 at the first day to
Arch Phys Med Rehabil Vol 91, July 2010
3.00 at the last day; P⬍.05), pain (massage group, from 4.11 at
baseline to 2.59 at 4 weeks of follow-up compared with con-
trols, from 6.17 at baseline to 4.83 at 4 weeks of follow-up;
P⬍.05), and grip strength (massage group, from 6.61 at base-
line to 7.8 at 4 weeks of follow-up compared with controls,
from 5.58 at baseline to 6.25 at 4 weeks of follow-up; P⬍.05)
after 3 treatment sessions at 4 weeks of follow-up.
Therefore, there is limited evidence that a targeted massage
protocol is more effective than a general massage protocol, and
that massage therapy for 15 minutes once a week with self-
massage daily is more effective than no treatment in the short
term.
Heat Wrap Therapy
Recent RCTs. One recent low-quality RCT60 (n⫽22) stud-
ied low-level heat wrap therapy (104°F; 40°C) for 3 days (with
a total of 26 time points) compared with oral placebo with a
follow-up of 2 days. Significant differences in favor of low-
level heat wrap therapy were found on pain at 20 of the 26 time
points (Pⱕ.05), joint stiffness reduction at 19 of the 26 time
points (Pⱕ.05), grip strength (mean ⫾ SD, heat wrap,
6.1⫾1.6kg, vs oral placebo, 0.8⫾1.4kg; P⫽.012) and symptom
severity scale (mean ⫾ SD, heat wrap, .97⫾.16, vs oral pla-
cebo, .14⫾.14; Pⱕ.001). After 3 days, significant differences
in favor of heat wrap therapy were found on function status
scale, but not at 5 days of follow-up (mean ⫾ SD, heat wrap,
.65⫾.16, vs oral placebo, .00⫾.16, P⫽.006, and heat wrap,
.57⫾.22, vs oral placebo, .12⫾.20, P⫽.07, respectively).
There is limited evidence that heat wrap therapy is more
effective than oral placebo in the short term (3 days of
follow-up).
Cupping Therapy
Recent RCTs. The high-quality study of Michalsen et al61
compared traditional cupping therapy with heat pads (control
group). At day 7, significant differences were found on pain at rest
(MD⫽⫺22.9; 95% CI, ⫺35.3 to ⫺10.5), the Levine CTS score
(symptom severity, mean difference, ⫺22.9, 95% CI, ⫺35.3 to
⫺10.5; functional status, MD ⫺0.6, 95% CI, ⫺0.8 to ⫺0.3), and
the DASH score (MD⫽⫺11.1; 95% CI, ⫺17.1 to ⫺5.1).
Therefore, we concluded that cupping therapy is more
effective (moderate evidence) than heat pads at 7 days of
follow-up.
Injections Other Than Steroids
Recent RCTs. An injection with botulinum B toxin into
each of the 3 hypothenar muscles was compared with placebo
in the low-quality study of Breuer et al62 (n⫽20). The study
reported no significant differences on Clinical Global Impres-
sion of Severity at 13 weeks of follow-up.
Thus, there is no evidence for the effectiveness of botulinum
B toxin compared with ibuprofen and wrist splint to treat
patients with CTS in the midterm.
Insulin as Additive to a Steroid Injection
Systematic review. The high-quality study of Ozkul et al63
investigated insulin as additive to steroid injection (methyl-
prednisolone 20mg in 1mL) for 7 weeks in patients with
noninsulin-dependent diabetes mellitus and found significant
differences on the Global Symptom scale in favor of steroid
injection plus insulin injections at 8 weeks of follow-up (no
exact data and P value given).
In conclusion, there is moderate evidence that in patients
with noninsulin-dependent diabetes mellitus, steroid injection
 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
plus insulin injections are more effective than steroid injections
alone for the treatment of CTS in the short term.
Ionthophoresis
Recent RCTs. One recent RCT of high quality64 found no
significant differences on the Levine Questionnaire between
dexamethasone iontophoresis and a control group (iontophore-
sis with distilled water) at 3 and 6 months of follow-up.
We concluded there is no evidence for the effectiveness of
dexamethasone iontophoresis compared with a placebo con-
trolled group in midterm and long term.
2. Corticosteroid injections
Marshall14 conducted a Cochrane review (search up to
May 2006) on local corticosteroid injection versus placebo
injection or other nonsurgical interventions in improving
clinical outcome and also to determine how long symptom
relief lasted. Twelve RCTs were included (n⫽671) in this
review. Furthermore, 3 recent RCTs were found.
Corticosteroid Injections Versus Placebo
Systematic review. One high-quality study65 (n⫽60) in-
cluded in the review of Marshall14 demonstrated significant
clinical improvement in favor of local corticosteroid (40mg
methylprednisolone) compared with placebo injection
(RR⫽3.83; 95% CI, 1.82– 8.05) 1 month after treatment.
Another high-quality study66 (n⫽81) compared 1.5mg beta-
methasone with placebo and found significant clinical improve-
ment in favor of corticosteroid injections 2 weeks after treat-
ment (RR⫽2.04; 95% CI, 1.26 –3.31). Pooling of the data of
the 2 RCTs demonstrated significant clinical improvement in
favor of corticosteroid injection in the short term (RR⫽2.58;
95% CI, 1.72–3.87).
In conclusion, we found strong evidence that a corticosteroid
injection is more effective than placebo in the treatment of
patients with CTS in the short term.
Local Versus Systemic Corticosteroid Injection
Systematic review. One high-quality trial67 (n⫽37)
showed a better rate of improvement with a local corticosteroid
injection (betamethasone 1.5mg) than with a systemic cortico-
steroid injection (betamethasone 1.5mg) (RR⫽3.17; 95% CI,
1.02–9.87) at 1 month of follow-up.
Therefore, there is moderate evidence that local corticoste-
roid injections are more effective than systemic corticosteroid
injections to treat CTS in the short term.
Corticosteroid Injection Versus Oral Steroid
Systematic review. One high-quality trial68 (n⫽60) in-
cluded in the Cochrane review14 found no significant differ-
ences on symptom improvement on the Global Symptom Score
at 2 weeks of follow-up and significant differences on symptom
improvement on the Global Symptom Score in favor of corti-
costeroid injections (15mg methylprednisolone) compared with
oral steroids (25mg methylprednisolone) at 8 weeks and 12
weeks of follow-up (WMD⫽⫺7.16, 95% CI, ⫺11.46 to
⫺2.86; and WMD⫽⫺7.10, 95% CI, ⫺11.68 to ⫺2.52,
respectively).
Recent RCTs. The long-term effects of the study of Wong
et al68 were reported by the high-quality study of Hui et al.69 Of
the 80 randomized participants, 35 did not require surgical
treatment in 80 weeks of follow-up; no significant differences
between these groups were found on the Global Symptom
Score at 80 weeks of follow-up (steroid injection, 69.5% im-
991
provement compared with baseline, vs oral prednisolone,
51.9% improvement compared with baseline; P⬎.05).
Thus, there is moderate evidence that corticosteroid injec-
tions are more effective than oral steroids in the short term.
Furthermore, there is no evidence for the effectiveness of
corticosteroid injections compared with oral steroids in the
treatment of patients with CTS in the long term.
Corticosteroid Injection Versus Anti-inflammatory
Medication Plus Splinting
Systematic review. In one high-quality trial70 (n⫽23) in-
cluded in the Cochrane review of Marshall,14 there was no
significant improvement in symptoms between the injection
group (40mg methylprednisolone) and the anti-inflammatory
medication (120mg acemetacin) plus splinting group at 2 and 8
weeks after treatment. Also, on pain (VAS), no significant
improvement was found at 2 and 8 weeks of follow-up.
We concluded that there is no evidence for the effectiveness of
corticosteroid injection compared with anti-inflammatory medica-
tion plus splinting as intervention for CTS in the short term.
Corticosteroid Injection Versus Helium-Neon Laser
Treatment
Systematic review. In the low-quality study of Lucantoni et
al71 (n⫽40), at 20 days of follow-up, significant differences
were found in favor of corticosteroid injections with 20mg
methylprednisolone compared with helium-neon laser on
symptom improvement (RR⫽1.89; 95% CI, 1.12–3.17). How-
ever, significant effects were no longer reported at 6 months of
follow-up.
Therefore, there is limited evidence that corticosteroid in-
jections are more effective than helium-neon laser in the short
term, but no evidence was found for the effectiveness in the
midterm.
Different Doses of Local Corticosteroid Injections
Systematic review. The low-quality study of O’Gradaigh
and Merry72 (n⫽64) found no significant differences on clin-
ical symptoms between the 25-mg hydrocortisone local injec-
tion group and the 100-mg hydrocortisone group at 6 weeks of
follow-up.
Recent RCTs. One high-quality RCT73 (n⫽172) reporting
on corticosteroid injections to treat CTS was found. At 1 year
of follow-up, better but nonsignificant differences in treatment
response were found for an injection with 60mg methylpred-
nisone compared with injections with 20mg or 40mg of the
same medication. At 6 months of follow-up, significantly better
results were found in favor of the 60-mg doses compared with
the other 2 doses (60-mg group, 73% [32/44] vs 40-mg group,
53% [23/43]) and 40mg (60-mg group, 73% [32/44] vs 20-mg
group, 56% [25/45]) of the same medication.
In conclusion, there is no evidence for the effectiveness of
25-mg hydrocortisone local injections compared with 100-mg
hydrocortisone injections in the short term. There is moderate
evidence that 60mg methylprednisone is more effective than 20
or 40mg methylprednisone in the midterm, but no evidence for
the effectiveness of 60mg methylprednisone compared with 20
or 40mg methylprednisone to treat CTS in the long term.
Short-Versus Long-Acting Corticosteroid Injection
Systematic review. The low-quality study of O’Gradaigh
and Merry72 (n⫽39) also examined the effectiveness of short-
acting local corticosteroid (100mg hydrocortisone) versus
long-acting corticosteroid (20mg triamcinolone). No signifi-
Arch Phys Med Rehabil Vol 91, July 2010
992 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
cant effects were found on clinical symptoms at 6 weeks of
follow-up.
Thus, there is no evidence for the effectiveness of short-
acting compared with long-acting corticosteroid injection in the
treatment of patients with CTS in the short term.
Single Versus 2 Local Corticosteroid Injections
Systematic review. One high-quality RCT74 (n⫽40) found
no significant differences on the Global Symptom Score be-
tween a single and 2 local corticosteroid injections (15mg
methylprednisolone) 8, 24, and 40 weeks after injection.
We concluded there is no evidence for the effectiveness of a
single compared with 2 local corticosteroid injections with
15mg methylprednisolone in the treatment of patients with
CTS in the short term, midterm, and long term.
Different Approaches of Corticosteroid Injection into the
Carpal Tunnel
Systematic review. One low-quality RCT75 (n⫽57) of the 2
RCTs included in the review of Marshall14 compared an injection
(3mg betamethasone disodium phosphate and 3mg betamethasone
acetate suspension mixed with 0.5cc lidocaine HCl 2% solution)
delivered 4cm proximal to the wrist flexor crease to an injection
more distal at the anterior wrist flexion crease. No significant
differences were found on the clinical Neurologic Symptom Score
at 11 months of follow-up. The low-quality study of Habib et al76
(n⫽42) examined a novel approach of local corticosteroid injec-
tions compared with the classic approach. In the novel approach,
patients were injected with 12mg methylprednisolone acetate us-
ing a 1-mL insulin syringe with a 29-gauge ⫻ 1/2-in ⫻ 3-cm
needle mixed with .15mL lidocaine 2%. The needle was totally
inserted (35° directing the needle toward the carpal tunnel) while
holding the hand in a midextended position. No significant differ-
ences were found between the 2 approaches on overall improve-
ment at 12 weeks of follow-up.
Therefore, there is no evidence for the effectiveness of a
novel approach compared with the classic approach in the short
term, or for a proximal compared with a distal approach of
corticosteroid injection to treat CTS in the long term.
Corticosteroid Injection Versus Iontoforese or
Phonophorese of Steroids
Systematic review. Two RCTs77,78 studied the efficacy of
local steroid injection compared with iontophoresis or phono-
phoresis. The low-quality study of Gökoğlu et al78 (n⫽30)
compared an injection with 40mg methylprednisone with ion-
tophoresis of dexamethasone (0.4%) and found significantly
better results in favor of the corticosteroid injection on pain
(WMD⫽⫺1.70; 95% CI, ⫺2.38 to ⫺1.02) and no significant
effect on the functional status scale and the symptom severity
scale at 8 weeks of follow-up. The low-quality study of Aygul
et al77 (n⫽21) found no significant differences between the
injection group (1mL dexamethasone) and the phonophoresis
group (dexamethasone 0.4%) on the symptom severity scale
and the functional status scale at 4 months of follow-up.
Thus, there is limited evidence that a corticosteroid injection
is more effective than iontophoresis in the short term, and no
evidence for the effectiveness of corticosteroid injection com-
pared with phonophoresis in the midterm.
Corticosteroid Injection Versus EMLA-crème
Recent RCTs. The low-quality study of Moghtaderi et al79
compared treatment with a steroid injection with 40mg meth-
ylprednisolone with the daily use of eutectic mixture of local
anesthetic (EMLA) cream and found significant differences on
Arch Phys Med Rehabil Vol 91, July 2010
pain at 4 weeks of follow-up within both groups (mean ⫾ SD,
EMLA, from 5.8⫾.98 at baseline to 2.1⫾1.2 at 4 weeks of
follow-up, P⬍.001; injection, from 5.7⫾1.0 at baseline to
1.6⫾1.4 at 4 weeks of follow-up, P⬍.001). No comparison
between the 2 groups was made.
We concluded that there is no evidence for the effectiveness of
1 steroid injection compared with EMLA cream in the short term.
DISCUSSION
The aim of this systematic review was to present an overview
of the current state of the art regarding the effectiveness of non-
surgical interventions for management of CTS. In comparison
with the conclusions of the Cochrane reviews of O’Connor13 and
Marshall,14 we found similar results for most interventions in-
cluded in these reviews. However, we also found benefit for
electromagnetic field therapy, ergonomic keyboards, cupping
therapy compared with heat pads in the short term, and ultrasound
in the midterm. Furthermore, our findings showed midterm— but
no long-term— benefit of steroid injections.
Strong and moderate evidence for effectiveness was found for
corticosteroids (oral or injected), and a corticosteroid injection
seems to be the most effective. In the midterm, moderate evidence
was found for a higher dose of injected methylprednisone (60mg)
compared with a lower dose (20 or 40mg). However, the included
studies that presented long-term results of steroids compared with
other interventions found that the positive results for the effec-
tiveness of steroids were not maintained in the long term. Similar
results were found in patients with other upper-extremity tendi-
nopathies, such as the trigger finger80 and lateral epicondylitis.81 It
seems that corticosteroid injections do not resolve the cause of
CTS but merely suppress its symptoms. The mechanism behind
this effect of corticosteroids in CTS remains unclear, but an
anti-inflammatory or a neovascular component may play a role. In
contrast, the finding that the effects of low and high doses do not
differ significantly in the long term raises the question whether the
effect is in fact a result of their pharmacologic properties. Alter-
native explanations are possible. However, caution is needed with
the use of corticosteroid injections because postinjection recur-
rence can increase, as reported in patients with lateral epicondy-
litis.82,83 Moreover, some authors suggest that potential harm
related to corticosteroid injections includes osteonecrosis and ten-
don rupture.84 In the review of Nichols,84 17 animal studies were
included. Ten of these studies provide evidence that corticosteroid
injections cause musculoskeletal structural or functional damage.
More studies are needed to elucidate the mechanism and effect of
corticosteroids in the longer term.
Ultrasound used as an intervention to treat CTS showed some
promising results. Moderate evidence was found for the effective-
ness of ultrasound in the short term and midterm. Further, we
found moderate evidence that ultrasound is more effective than
laser therapy in the short term. Ultrasound has a long history of
use in physiotherapy practice.85 It is assumed to increase temper-
ature in deep tissue by increasing blood flow, tissue metabolism,
and nerve function.86-88 Nerve generation can indeed be influ-
enced by temperature.89 Further, in patients with peripheral neu-
ropathy, a reversible conduction block as a result of acute ultra-
sound may occur.90 According to Watson,85 the clinical outcome
of ultrasound appears to be dose-dependent. The studies we in-
cluded that reported on the effectiveness of different intensities36
or frequencies37 of ultrasound found no evidence for this state-
ment; however, they only studied results at 2 and 4 weeks,
respectively. More RCTs are needed to establish the efficacy and
safety of ultrasound in the short term, midterm, and long term in
patients with CTS.
Moderate evidence was found for the effectiveness of dynamic
magnetic field therapy in the short term. This conclusion was
 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede 993

made on the basis of 1 small (n⫽36) high-quality study. Although
no significant differences were found on pain measured on a VAS,
significantly better results were found on the Neuropathy Pain
Scale, with a pain reduction of 42% in the treatment group
compared with 24% in the control group. More research is needed
to confirm these findings and explore the effect of this treatment in
the midterm and long term.
Splinting is a commonly used nonsurgical intervention to treat
CTS.10,91 We found moderate evidence for the effectiveness of the
use of a nocturnal hand brace compared with no treatment in the
short term. Further, we found that night splinting is as effective as
whole-day splinting. This may suggest that a patient can wear a
splint only at night to achieve a similar improvement in outcome.
However, the RCTs we found studied splinting in the short term
and midterm only. No long-term results for the effectiveness of
splinting were found.
Further, moderate evidence was found for the effectiveness
of an ergonomic keyboard compared with a standard keyboard
in the short term. Typing is an activity involving repetition,
vibration, and sustained posture. It is known that these move-
ments may contribute to symptoms of work-related upper ex-
tremity disorders in general92 and to CTS in particular.7 Our
findings contribute to this statement for patients with CTS.
Study Limitations
Some limitations of this review and its conclusions should
be addressed. First, we refrained from statistical pooling of the
results of the individual trials. This was done because of the
perceived heterogeneity of the studies. A single-point estimate of
the effect of the interventions included for CTS would probably
not do justice to the differences among the trials regarding patient
characteristics, interventions, and outcome measures. The use of a
best-evidence synthesis is a next-best solution and a transparent
method that is commonly applied in the field of musculoskeletal
disorders when statistical pooling is not feasible or clinically
viable.16
Second, only 55% of the included RCTs were of high quality.
Therefore, more high-quality RCTs are needed to obtain evidence-
based results with a low risk of bias. We used the list of Furlan15
that was constructed by the Cochrane Back Review Group. This
list was constructed to assess interventions in the field of neck and
back disorders but can also be used in and appears very suitable to
other fields.93,94 Furlan15 indicated that studies with serious flaws
or those in which fewer than 6 of the criteria are met should be
rated as having a high risk of bias. Therefore, these studies are
indicated as low-quality studies in our review. Although selecting
a border between high and low quality is arbitrary, there is em-
pirical evidence from a methodologic study conducted with data
from the Cochrane Collaboration Back Review Group that a
threshold of fewer than 50% of the criteria is associated with
bias.95
Third, the included Cochrane reviews of O’Connor13 and Mar-
shall14 used different methodologic quality criteria compared with
our criteria based on Furlan et al.96 Because of the high credibility
and validity of Cochrane reviews, we decided to apply the meth-
odologic quality criteria and definitions of high-quality and low-
quality studies used in a Cochrane review. However, the quality
criteria of the 2 Cochrane reviews included fewer items than our
12 quality criteria. This could contribute to bias in outcome of
evidence and conclusions. Further, O’Connor13 used an A (ie,
high quality: all criteria met), a B (ie, moderate quality: 1 or more
criteria partly met) and a C (ie, low quality: 1 or more criteria not
met) as methodologic quality scores. For our review, we decided
to rank A and B as high-quality and C as low-quality studies.
However, we also looked at the number of items that were scored
positive. Based on our definition and use of high quality (ie, 50%
or more of the items were scored positive), we found a discrep-
ancy in 3 of the studies: Oztas et al36 scored 3 out of 8 items but
it was ranked as B (ie, of high quality). Out of 8 items, Davis51 and
Manente et al23 scored 5 and 4 items positively, respectively, but
both were ranked as C (ie, of low quality). This could introduce
potential bias. However, even if we had changed the rankings of
these 3 studies, our additional analysis shows that our study
conclusions would not have changed.
CONCLUSIONS
In conclusion, strong and moderate evidence was found for the
effectiveness of interventions to treat CTS with oral steroids,
steroid injections, ultrasound, electromagnetic field therapy, noc-
turnal splinting, ergonomic keyboards compared with a standard
keyboard, and cupping therapy compared with heat pads in the
short term. Also, moderate evidence was found for ultrasound in
the midterm. With the exception oral or injected steroids, no
long-term results were reported for any of these treatments. More-
over, although a higher dose of steroid injections seems to be more
effective in the midterm, the benefits of steroids injections were
not maintained in the long term.

APPENDIX 1: SEARCH STRINGS

PubMed
CTS
“Carpal tunnel syndrome”[mh] OR (“median nerve”[mh] AND (compress* OR entrapment)) OR “carpal tunnel” OR ((median*
AND (nervus OR nerve)) AND (compress* OR entrapment))Therapy
(randomized controlled trial[Publication Type] OR (randomized[Title/Abstract] AND controlled[Title/Abstract] AND trial[Title/
Abstract]))Systematic Reviews
((meta-analysis [pt] OR meta-analysis [tw] OR metanalysis [tw]) OR ((review [pt] OR guideline [pt] OR consensus [ti] OR
guideline* [ti] OR literature [ti] OR overview [ti] OR review [ti]) AND ((Cochrane [tw] OR Medline [tw] OR CINAHL [tw] OR
(National [tw] AND Library [tw])) OR (handsearch* [tw] OR search* [tw] OR searching [tw]) AND (hand [tw] OR manual [tw]
OR electronic [tw] OR bibliography* [tw] OR database* OR (Cochrane [tw] OR Medline [tw] OR CINAHL [tw] OR (National
[tw] AND Library [tw]))))) OR ((synthesis [ti] OR overview [ti] OR review [ti] OR survey [ti]) AND (systematic [ti] OR critical
[ti] OR methodologic [ti] OR quantitative [ti] OR qualitative [ti] OR literature [ti] OR evidence [ti] OR evidence-based [ti])))
BUTNOT (case* [ti] OR report [ti] OR editorial [pt] OR comment [pt] OR letter [pt])RCTs
(randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [mh] OR random allocation
[mh] OR double-blind method [mh] OR single-blind method [mh] OR clinical trial [pt] OR clinical trials [mh] OR “clinical trial”
[tw] OR ((singl* [tw] OR doubl* [tw] OR trebl* [tw] OR tripl* [tw]) AND (mask* [tw] OR blind* [tw])) OR “latin square” [tw]
OR placebos [mh] OR placebo* [tw] OR random* [tw] OR research design [mh:noexp] OR comparative study [mh] OR evaluation

Arch Phys Med Rehabil Vol 91, July 2010

994 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede

studies [mh] OR follow-up studies [mh] OR prospective studies [mh] OR cross-over studies [mh] OR control* [tw] OR
prospective* [tw] OR volunteer* [tw]) NOT (animal [mh] NOT human [mh])Limits Activated
Humans, English, French, German, Dutch

EMBASE
CTS
“carpal tunnel syndrome”/ OR (“median nerve”/ AND “nerve compression”/) OR (“median nerve”/ AND (compress* OR
entrapment)):ti,ab OR “carpal tunnel” OR ((median* AND (nervus OR nerve)) AND (compress* OR entrapment))Therapy
“randomized controlled trial”:it OR (randomized:ti,ab AND controlled:ti,ab AND trial:ti,ab)Systematic Reviews
(“review”/exp AND (medline:ti,ab OR medlars:ti,ab OR embase:ti,ab OR pubmed:ti,ab) OR scisearch:ti,ab OR psychlit:ti,ab OR
psyclit:ti,ab OR psycinfo:ti,ab OR pyschinfo:ti,ab OR cinahl:ti,ab OR “hand search”:ti,ab OR “manual search”:ti,ab OR “electric
database”:ti,ab OR “bibliographic database”:ti,ab OR “pooled analysis”:ti,ab OR “pooled analyses”:ti,ab OR pooling:ti,ab OR
peto:ti,ab OR dersimonian:ti,ab OR “fixed effect”:ti,ab OR “mantel haenszel”:ti,ab OR “retracted article”:ti,ab) OR (“meta
analysis”/exp OR “meta analysis” OR “meta-analysis” OR “meta-analyses”:ti,ab OR “meta analyses”:ti,ab OR “systematic
review”:ti,ab OR “systematic overview”:ti,ab OR “quantitative review”:ti,ab OR “quantitative overview”:ti,ab OR “methodologic
review”:ti,ab OR “methodologic overview”:ti,ab OR “integrative research review”:ti,ab OR “research integration”:ti,ab OR
“quantitative synthesis”:ti,ab)RCTs
(“controlled clinical trial”/exp OR “randomized controlled trial”:ti OR “controlled clinical trial”:it OR “randomization”/ OR
“double blind procedure”/ OR “single blind procedure”/ OR “crossover procedure”/ OR “clinical trial”:it OR ((“clinical trial” OR
(singl* OR doubl* OR tripl*)) AND (mask* OR blind*)) OR (“Latin square design”/ OR “latin square” OR “latin-square”) OR
“placebo”/ OR placebo* OR “random sample”/ OR “comparative study”:it OR “evaluation study”:it OR evaluation/exp OR “follow
up”/exp OR “prospective study”/ OR control* OR prospective* OR volunteer*) NOT (animals/exp NOT humans/exp)Limits
Activated
Humans, English, French, German, Dutch

CINAHL
CTS
(MH “Carpal tunnel syndrome”) or ((MH “median nerve”) and (compress* or entrapment)) or “carpal tunnel” or ((median* and
nerv*) and (compress* or entrapment))Reviews
(MH “Systematic Review”)Clinical Trials
(MH “Clinical Trials⫹”)

PEDro
Carpal tunnel syndrome

APPENDIX 2: OPERATIONALIZATION CRITERIA METHODOLOGIC QUALITY

Criteria for a judgment of “yes” for the sources of risk of bias
1. Was the method of randomization adequate?
A random (unpredictable) assignment sequence. Examples of adequate methods are coin toss (for studies with 2 groups), rolling
a dice (for studies with 2 or more groups), drawing of balls of different colors, drawing of ballots with the study group labels from
a dark bag, computer-generated random sequence, preordered sealed envelopes, sequentially-ordered vials, telephone call to a
central office, and preordered list of treatment assignments.
Examples of inadequate methods are alternation, birth date, social insurance/security number, date in which they are invited to
participate in the study, and hospital registration number.
2. Was the treatment allocation concealed?
Assignment generated by an independent person not responsible for determining the eligibility of the patients. This person has
no information about the persons included in the trial and has no influence on the assignment sequence or on the decision about
eligibility of the patient.
Was knowledge of the allocated interventions adequately prevented during the study?
3. Was the patient blinded to the intervention?
This item should be scored “yes” if the index and control groups are indistinguishable for the patients or if the success of blinding
was tested among the patients and it was successful.
4. Was the care provider blinded to the intervention?
This item should be scored “yes” if the index and control groups are indistinguishable for the care providers or if the success
of blinding was tested among the care providers and it was successful.
5. Was the outcome assessor blinded to the intervention?
Adequacy of blinding should be assessed for the primary outcomes. This item should be scored “yes” if the success of blinding
was tested among the outcome assessors and it was successful or:
● For patient-reported outcomes in which the patient is the outcome assessor (eg, pain, disability): the blinding procedure is
adequate for outcome assessors if participant blinding is scored “yes.”
● For outcome criteria assessed during scheduled visit and that supposes a contact between participants and outcome assessors
(eg, clinical examination): the blinding procedure is adequate if patients are blinded and the treatment or adverse effects of
the treatment cannot be noticed during clinical examination.

Arch Phys Med Rehabil Vol 91, July 2010

 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede 995

● For outcome criteria that do not suppose a contact with participants (eg, radiography, magnetic resonance imaging): the
blinding procedure is adequate if the treatment or adverse effects of the treatment cannot be noticed when assessing the main
outcome.
● For outcome criteria that are clinical or therapeutic events that will be determined by the interaction between patients and care
providers (eg, co-interventions, hospitalization length, treatment failure), in which the care provider is the outcome assessor:
the blinding procedure is adequate for outcome assessors if item “E” is scored “yes.”
● For outcome criteria that are assessed from data of the medical forms: the blinding procedure is adequate if the treatment or
adverse effects of the treatment cannot be noticed on the extracted data.
Were incomplete outcome data adequately addressed?
6. Was the dropout rate described and acceptable?
The number of participants who were included in the study but did not complete the observation period or were not included
in the analysis must be described and reasons given. If the percentage of withdrawals and dropouts does not exceed 20% for
short-term follow-up and 30% for long-term follow-up and does not lead to substantial bias, a “yes” is scored. (N.B. these
percentages are arbitrary, not supported by literature).
7. Were all randomized participants analyzed in the group to which they were allocated?
All randomized patients are reported/analyzed in the group they were allocated to by randomization for the most important
moments of effect measurement (minus missing values) irrespective of noncompliance and co-interventions.
8. Are reports of the study free of suggestion of selective outcome reporting?
In order to receive a “yes,” the review author determines whether all the results from all prespecified outcomes have been
adequately reported in the published report of the trial. This information is either obtained by comparing the protocol and the report,
or in the absence of the protocol, assessing that the published report includes enough information to make this judgment.
Other sources of potential bias
9. Were the groups similar at baseline regarding the most important prognostic indicators?
In order to receive a “yes,” groups have to be similar at baseline regarding demographic factors, duration and severity of
complaints, percentage of patients with neurological symptoms, and value of main outcome measures.
10. Were co-interventions avoided or similar?
This item should be scored “yes” if there were no co-interventions or they were similar between the index and control groups.
11. Was the compliance acceptable in all groups?
The reviewer determines whether the compliance with the interventions is acceptable, based on the reported intensity, duration,
number, and frequency of sessions for both the index intervention and control interventions. For example, physiotherapy treatment
is usually administered over several sessions; therefore, it is necessary to assess how many sessions each patient attended. For
single-session interventions (for example, surgery), this item is irrelevant.
12. Was the timing of the outcome assessment similar in all groups?
Timing of outcome assessment should be identical for all intervention groups and for all important outcome assessments.

APPENDIX 3: SYSTEMATIC REVIEWS: CTS

No. of
Author Patients Treatment Placebo Control/Comparison Outcome Measures Effect Size

Nonsurgical treatment
other than injection
Splints
O’Connor 923 Splint in neutral Splint 20° extension Symptom relief overall 2wk RR⫽2.43 (95% ci, 1.12 to 5.28)
et al13 position In favor of splint in neutral position
(21 RCTs) Symptom relief at night Symptom relief at night
2wk RR⫽2.14 (95% ci, .99 to 4.65)
In favor of splint in neutral position

Nocturnal brace No treatment Symptom improvement 2wk WMD⫽⫺1.03 (95% CI, ⫺1.31 to ⫺.75)
In favor of nocturnal brace
4wk wmd⫽⫺1.07 (95% CI, ⫺1.29 to ⫺.85)
In favor of nocturnal brace
Hand function 2wk WMD⫽⫺.52 (95% CI, ⫺.79 to ⫺.25)
In favor of nocturnal brace
4wk WMD⫽⫺.55 (95% CI, ⫺.82 to ⫺.28)
In favor of nocturnal brace
Overall improvement 4wk RR⫽4.00 (95% CI, 2.34 to 6.84)
In favor of nocturnal brace

Full-time splint
Nerve/tendon gliding
Exercises and splint
Nighttime splint Symptom improvement WMD⫽⫺.21 (95% ci, ⫺.83 to .41)
Hand function WMD⫽⫺.21 (95% CI, ⫺.87 to .45)
Splint only Symptom improvement 3mo WMD⫽⫺3.68 (95% CI, ⫺8.56 to 1.20)
Hand function 3mo WMD⫽⫺1.00 (95% CI, ⫺4.72 to 2.72)
Grip strength 3mo WMD⫽5.06 (95% CI, ⫺5.51 to 15.63)
Pinch strength 3mo WMD⫽5.27 (95% CI, ⫺.94 to 11.48)

Neurodynamic Control group Improved pain RR⫽15.00 (95% CI, 1.02 to 220.92)
Mobilization hand function RR⫽9.00 (95% CI, .59 to 137.65)
Wrist flexion WMD⫽7.28 (95% CI, ⫺3.33 to 17.89)
Wrist extension WMD⫽6.00 (95% CI, ⫺4.56 to 16.56)
symptom improvement WMD⫽⫺.57 (95% CI, ⫺1.73 to .59)

Arch Phys Med Rehabil Vol 91, July 2010

996 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede

APPENDIX 3: SYSTEMATIC REVIEWS: CTS (Cont’d)

No. of
Author Patients Treatment Placebo Control/Comparison Outcome Measures Effect Size

Yoga Splint Pain (VAS) 8wk WMD⫽⫺1.40 (95% CI, ⫺2.73 to ⫺.07)
Grip strength 8wk WMD⫽⫺3.10 (95% CI, ⫺44.57 to 38.37)
Ultrasound
Pulsed ultrasound x Symptom improvement 2wk WMD⫽⫺.11 (95% CI, ⫺.67 to .45)
therapy 7wk WMD⫽⫺.99 (95% CI, ⫺1.77 to ⫺.21)
In favor of ultrasound
6mo WMD⫽⫺1.86 (95% CI, ⫺2.67 to ⫺1.05)
In favor of ultrasound
Pain (VAS) 6mo WMD⫽⫺1.10 (95% CI, ⫺2.92 to .72)
Grip strength 6mo WMD⫽4.16 (95% CI, ⫺.88 to 9.20)
Pinch strength 6mo WMD⫽.74 (95% CI, ⫺.17 to 1.65)
Self-reported 6mo RR⫽1.91 (95% CI, 1.13 to 3.23)
improvement

Continuous ultrasound x Pain (VAS) WMD⫽⫺.70 (95% CI, ⫺2.28 to .88)

therapy of 2 Symptom improvement WMD⫽⫺.30 (95% CI, ⫺.90 to .30)
different intensities,
1.5W and 0.8W

Ultrasound therapy Ultrasound therapy Improved pain RR⫽.63 (95% CI, .26 to 1.52)
1MHz 3MHz
Oral medication or
vitamins
Diuretics x Symptom improvement 4wk WMD⫽.80 (95% CI, ⫺3.67 to 5.27)
(trichlormethiazide 6mo RR⫽.98 (95% CI, .68 to 1.42)
or bendrofluazide

NSAIDs (tenoxicam) x Symptom improvement 2wk WMD⫽3.10 (95% CI, ⫺1.96 to 8.16)
4wk WMD⫽3.20 (95% CI, ⫺2.33 to 8.73)

Oral steroids x Symptom improvement 2wk WMD⫽⫺7.23 (95% CI, ⫺10.31 to ⫺4.14)
(prednisone or In favor of oral steroids
prednisolone) 4wk WMD⫽⫺10.8 (95% CI, ⫺15.26 to ⫺6.34)
In favor of oral steroids
8wk WMD⫽⫺6.46 (95% CI, ⫺11.93 to ⫺.99)

Diuretics NSAIDs (tenoxicam) Symptom improvement 2wk WMD⫽2.40 (95% CI, ⫺7.20 to 2.40)
(trichlormethiazide) 4wk WMD⫽⫺2.40 (95% CI, ⫺7.20 ⫺ 2.40)

Diuretics Oral steroids Symptom improvement 2wk WMD⫽7.30 (95% CI, 3.43 to 11.17)
(trichlormethiazide) (prednisolone) 4wk WMD⫽11.60 (95% CI, 7.25 to 15.95)

NSAIDs (tenoxicam) Oral steroids Symptom improvement 2wk WMD⫽9.70 (95% CI, 4.85 to 14.55)
(prednisolone) 4wk WMD⫽14.00 (95% CI, 8.57 to 19.43)

Pyridoxine (B6) 200mg x Nocturnal discomfort 10–12wk WMD⫽⫺.50 (95% CI, ⫺1.37 to .37)
Hand coordination 10–12wk WMD⫽23⬎.60 (95% CI, ⫺1.57 to .37)
Symptom improvement 10–12wk RR⫽.63 (95% CI, .25 to 1.56)
Mobilization
Carpal bone Control group Improved pain 3wk RR⫽15.00 (95% CI, 1.02 to 220.92)
mobilization Hand function 3wk RR⫽11.00 (95% CI, .74 to 163.49)
Wrist flexion 3wk WMD⫽6.43 (95% CI, ⫺4.50 to 17.36)
Wrist extension 3wk WMD⫽6.86 (95% CI, ⫺1.90 to 15.62)
Symptom improvement 3wk WMD⫽⫺1.43 (95% CI, ⫺2.19 to ⫺.67)

Neurodynamic Carpal bone Improved pain Not estimable

mobilization Hand function RR⫽.80 (95% CI, .41 to 1.56)
Wrist flexion WMD⫽.85 (95% CI, ⫺10.83 to 12.53)
Wrist extension WMD⫽⫺.86 (95% CI, ⫺9.26 to 7.54)
Symptom improvement WMD⫽.86 (95% CI, ⫺.32 to 2.04)
Improved grasp RR⫽3.27 (95% CI, .15 to 72.23)

Ergonomic keyboards
Ergonomic keyboards
Standard keyboard Pain (VAS) 12wk WMD⫽⫺2.40 (95% CI, ⫺4.45 to ⫺.35)
In favor of ergonomic keyboard
Hand function 12wk WMD⫽⫺2.20 (95% CI, ⫺12.08 to 7.68)
In favor of ergonomic keyboard

Comfort keyboard Regular keyboard Pain (VAS) 6mo WMD⫽.97 (95% CI, ⫺.64 to 2.58)
system Hand function 6mo WMD⫽.57 (95% CI, ⫺.15 to 1.29)

Apple adjustable Regular keyboard Pain (VAS) 6mo WMD⫽.70 (95% CI, ⫺.97 to 2.37)
keyboard Hand function 6mo WMD⫽.93 (95% CI, .26 to 1.60)
In favor of apple keyboard

Arch Phys Med Rehabil Vol 91, July 2010

 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede 997

APPENDIX 3: SYSTEMATIC REVIEWS: CTS (Cont’d)

No. of
Author Patients Treatment Placebo Control/Comparison Outcome Measures Effect Size

Microsoft natural Regular keyboard Pain (VAS) 6mo WMD⫽.79 (95% CI, ⫺1.53 to 3.11)
keyboard Hand function 6mo WMD⫽1.92 (95% CI, .84 to 3.00)
In favor of Microsoft keyboard
Magnet therapy
Magnet therapy x Improved pain 2wk WMD⫽.00 (95% CI, ⫺2.82 to 2.82)

Acupuncture
Laser acupuncture x Night pain 3wk RR⫽1.32 (95% CI, .93 to 1.86)

Chiropractic therapy
Chiropractic treatment Ibuprofen and splint Hand function WMD⫽⫺3.30 (95% CI, ⫺9.74 to 3.14)

Injections
Steroid injection Steroid injection Global Symptom Score 8wk: significant in favor of steroid plus
(methylprednisolone (methylprednisolone insulin injections (no exact data given)
20mg in 1 ml) 20mg in 1 mL)
followed by weekly followed by placebo
injections of injections
NPH insulin
(0.3mL – 12U)

Local corticosteroid
injection
Marshall 671 40mg x Clinical improvement 1mo RR⫽3.83 (95% CI, 1.82 to 8.05)
et al14 methylprednisolone In favor of methylprednisolone
(12 RCTs) and 10mg lidocaine

1mL betamethasone x Clinical improvement 2wk RR⫽2.04 (95% CI, 1.26 to 3.31)
In favor of betamethasone

Corticosteroids x Clinical improvement RR⫽2.58 (95% CI, 1.72 to 3.87)

1.5mg betamethasone Systemic injection of Clinical improvement RR⫽3.17 (95% CI, 1.02 to 9.87)
1.5mg In favor of local injection
betamethasone

15mg 25mg Clinical improvement, 2wk WMD⫽⫺4.20 (95% CI, ⫺8.66 to .26)
methylprednisolone methylprednisolone Global Symptom 8wk WMD⫽⫺7.16 (95% CI, ⫺11.46 to ⫺2.86)
oral Score In favor of corticosteroid injection
12wk WMD⫽⫺7.10 (95% CI, ⫺11.68 to ⫺2.52)
in favor of corticosteroid injection

40mg 120mg acemetacine Symptom severity scale 2wk WMD⫽.00 (95% CI, ⫺.64 to .64)
methylprednisolone and night splint 8wk WMD⫽.10 (95% CI, ⫺.33 to .53)
Pain (VAS) 2wk WMD⫽⫺1.20 (95% CI, ⫺2.71 to .31)
8wk WMD⫽.10 (95% CI, ⫺1.13 to 1.33)

20mg Helium neon laser Symptom improvement 20d RR⫽1.89 (95% CI, 1.12 to 3.17)
methylprednisolone In favor of corticosteroid injection
6mo MD⫽.75 (95% CI, ⫺2.81 to 4.31)

35mg
methylprednisolone
acetate 0.5mL and
lidocaine 2% at the
wrist crease
15mg Clinical improvement 12wk RR⫽1.00 (95% CI, .59 to 1.69)
methylprednisolone
acetate and .15mL
lidocaine 2% at a
location 2–3cm
distal to the wrist
crease

25mg hydrocortisone 100mg hydrocortisone Clinical symptoms 6wk RR⫽1.05 (95% CI, .73 to 1.52)

100mg hydrocortisone 20mg triamcinolone Clinical symptoms 6wk RR⫽1.08 (95% CI, .71 to 1.64)

15mg 15mg Global Symptom Score 8wk MD⫽⫺3.80 (95% CI, ⫺9.27 to 1.67)
methylprednisolone methylprednisolone Symptom Score 24wk MD⫽⫺2.90 (95% CI, ⫺9.20 to 3.40)
2 times injection 1 time distal Clinical Neurologic 40wk MD⫽1.50 (95% CI, ⫺4.76 to 7.76)
4cm proximal to the injection at the 11mo WMD⫽2.17 (95% CI, ⫺1.07 to 5.41)
wrist flexor crease anterior wrist flexion
(3mg crease (3mg
betamethasone betamethasone
disodium disodium
phosphate, 3 mg phosphate, 3mg
betamethasone betamethasone
acetate suspension, acetate suspension,
0.5 cc lidocaine HCl 0.5cc lidocaine HCl
2% solution) 2% solution)

Arch Phys Med Rehabil Vol 91, July 2010

998 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede

APPENDIX 3: SYSTEMATIC REVIEWS: CTS (Cont’d)

No. of
Author Patients Treatment Placebo Control/Comparison Outcome Measures Effect Size

40mg Iontophoresis of Pain (VAS) 8wk WMD⫽⫺1.70 (95% CI, ⫺2.38 to ⫺1.02)
methylprednisone dexamethasone Functional status scale 8wk WMD⫽⫺.28 (95% CI, ⫺.95 to .39)
0.4% Symptom Severity 8wk WMD⫽⫺.29 (95% CI, ⫺.63 to .05)
Scale

1mL dexamethasone 3MHz and intensity of Symptom severity scale 4mo WMD⫽⫺.40 (95% CI, ⫺.93 to .13)
1.0W/cm2 and 0.1% Functional status scale 4mo WMD⫽⫺.17 (95% CI, ⫺.53 to .19)
dexamethasone

Abbreviations: x, studies in which the intervention group was compared to a placebo group; NSAIDs, nonsteroidal anti-inflammatory drugs; NPH, Neutral Protamine Hagedorn.

APPENDIX 4: RECENT RCTs: CTS

Outcome Measures Results:

Author Treatment Placebo Control/Comparison (Total Follow-Up Time) Statistical Results by Outcome Measure

>Nonsurgical treatment
(other than injection)
Splinting
Premoselli nocturnal neutral wrist Mo treatment Levine’s Questionnaire:
et al29 splint for 6mo (n⫽25) (n⫽25) symptoms P⫽.001 3mo differences:
(6mo) mean ⫾ SD, splint,* 1.07⫾.39, vs control, ⫺.02⫾.24
P⫽.001 6mo differences:
splint,* 1.22⫾.39, vs control, .17⫾.29
Functions P⫽.0001 3mo differences:
(6mo) mean ⫾ SD, splint,* .53⫾.22, vs control, ⫺.15⫾.43
P⫽.0004 6mo differences:
splint,* .75⫾.28, vs control, .04⫾.30

De Angelis Wrist splint “CAMP TIELLE” Hand brace Pain (VAS) Comparison between groups: differences from
et al31 at night for 3mo (n⫽61) “MANU” at night (9mo) baseline
for 3mo (n⫽59) P⫽.647 to 3mo follow-up: mean (95% CI), .10
(⫺7.10 to 11.36)
P⫽.380 to 9mo follow-up: 3.10 (⫺6.42 to 16.67)
Boston Carpal Tunnel P⫽.556 Comparison between groups: differences from
Questionnaire baseline
Symptom severity to 3mo follow-up: mean (95% CI), ⫺.06 (⫺.34 to
score (9mo) .18)
P⫽.778 to 9mo follow-up: .07 (⫺.28 to .36)
Function severity score P⫽.471 Comparison between groups: differences from
(9mo) baseline
to 3mo follow-up: mean (95% CI), .19 (⫺.16 to .34)
P⫽.656 to 9mo follow-up: .17 (⫺.22 to .35)
Brininger Group 1: neutral wrist and Group 3: idem plus Symptoms (8wk) Significant, no P No occasional symptoms: group 1: 38% vs group
et al26 MCP splint (n⫽13) tendon and nerve Pinch strength (4wk) value given Not 2: 17%†
group 2: wrist cock-up group 1 plus tendon significant group 3 and group 4†
splint (n⫽14) and nerve gliding No P value all groups†
All groups had a treatment exercises (n⫽13) given
periof of 4k group 4: idem Significant, no P
group 2 plus tendon value given
and nerve gliding
exercises (n⫽11)
Pinar Nerve gliding exercises for Nerve gliding Pain value (VAS) P⬎.05 Mean ⫾ SD, 6.9⫾1.4 to 1.0⫾1.6 in experimental
et al28 10wk plus splint 10 weeks exercises for 6wk (10wk) group vs 6.9⫾1.5 to 1.6⫾1.8 in control group
(n⫽19 hands) plus splint 10wk
(n⫽16 hands)
Grip strength (kg) P⬍.05 in favor of Mean ⫾ SD, 17.8⫾6.1 to 22.0⫾6.8 in experimental
(10wk) experimental group* vs 20.4⫾4.7 to 21.7⫾4.3 in control group
group
Pinch strength (kg) P⬎.05 Mean ⫾ SD, 4.1⫾1.7 to 5.4⫾1.8 in experimental
(10wk) group vs 4.3⫾1.3 to 4.9⫾1.1 in control group
Heebner Standard care plus active Standard care (ie, DASH Questionnaire P⫽.164 No exact data given
and neurodynamic exercises splinting at night (6mo)
Roddey27 (3–5 times/d; group 1) and at heavy
(n⫽30) activities, tendon
gliding exercises
3–5 times/d; group
2; n⫽30)
The Brigham and P⫽.080 No exact data given
Women’s Hospital
Carpal Tunnel Specific
Questionnaire:
Symptom severity
score (6mo)

Arch Phys Med Rehabil Vol 91, July 2010

 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede 999

APPENDIX 4: RECENT RCTs: CTS (Cont’d)

Outcome Measures Results:
Author Treatment Placebo Control/Comparison (Total Follow-Up Time) Statistical Results by Outcome Measure

Function severity score P⫽.016 Group 1*: mean, 2.2, vs group 2, 2.9
(6mo)
Neurodynamic P⫽.336 No exact data given
irritability of the
median nerve
Baysal 1. A custom-made neutral 2. Splinting with Pain (VAS) (8wk) P⬍.05 Group 1: mean ⫾ SD, 4.8⫾2.3 before treatment to
et al25 volar splint for 3wk with ultrasound (all groups) 2.6⫾ 2.8 after 8wk; group 2: 5.7⫾2.7 to 2.5⫾2.8;
nerve and tendon gliding treatment 15 min/ Hand grip strength group 3: 5.6⫾3.5 to 0.8⫾0.9†
exercises, 5 sessions daily session to the (8wk) P⬍.05 Group 1: mean ⫾ SD, 20.6⫾7.1 before treatment to
for 3wk (n⫽24) palmar carpal (all groups) 22.7⫾7.4 after 8wk; group 2: 20.6⫾10.1 to 23.5⫾
tunnel, 5 times/wk 2.6; group 3: 20.7⫾5.5 to 22.3⫾5.1†
for 3 wk (n⫽16)
3. Splinting with Pinch strength (8wk) P⬍.05 (all Group 1: mean ⫾ SD, 4.9⫾2.5 before treatment to
ultrasound groups) 6.3⫾ 1.7 after 8wk; group 2: 4.3⫾2.2 to 5.7⫾2.3;
treatment with group 3: 5.6⫾1.4 to 7.0⫾2.2†
nerve and tendon
gliding exercises
(n⫽16)
Levine’s Questionnaire P⬍.05 (all Group 1: mean ⫾ SD, 28⫾9.7 before treatment to
Symptom severity groups) 20.2⫾ 10.4 after 8wk; group 2: 29.6⫾9.7 to
scale (8wk) 19.1⫾9.4; group 3: 30.4⫾12.1 to 15.6⫾4.7(†)
Function status scale P⬍.05 (all Group 1: mean ⫾ SD, 20.6⫾7.8 before treatment to
(8wk) groups) 14.9⫾ 6.6 after 8wk; group 2: 21.9⫾9.1 to 16.1⫾8.7;
group 3: 20.5⫾7.1 to 12.6⫾3.4†
Mishra Splint in neutral position for Oral prednisolone Boston Carpal Tunnel P⫽.03 Difference of mean between baseline and 3mo
et al33 4wk (n⫽36) 20 mg/d for 2wk Questionnaire Function follow-up: mean ⫾ SD, splint, .16⫾.17, vs oral
followed by 10mg/d status score (3mo) steroids,* .26⫾.21
for 2wk (n⫽35)
Symptom severity P⫽.42 Difference of mean between baseline and 3mo
score (3mo) follow-up: mean ⫾ SD, splint, .39⫾.54, vs oral
steroids, .49⫾.44
Yagci Full-time hand splint in Full-time hand Boston Carpal Tunnel P⫽.42 Mean ⫾ SD, splint: from 2.49⫾.65 at baseline to
et al30 neutral position for 3 mo splint in neutral Questionnaire Function 2.38⫾ .71 at 3mo follow-up
(n⫽24) position plus 10 capacity score (3mo)
sessions of low-
level laser therapy
(n⫽21)
P⫽.21 Splint plus laser: from 2.39⫾1.03 at baseline to
2.1⫾.63 at 3mo follow-up
Not significant Splint vs splint plus laser
No P value
given
Symptom severity P⫽.001 Mean ⫾ SD, splint: from 2.91⫾.64 at baseline to
score (3mo) 2.35⫾.65 at 3mo follow-up vs
P⫽.044 Splint plus laser: from 2.50⫾.79 at baseline to
2.25⫾.792 at 3mo follow-up
Not significant Splint vs splint plus laser
No P value
given
Grip strength P⫽.016 Mean ⫾ SD, splint: from 29.11⫾7.28 at baseline to
26.83⫾7.16 at 3mo follow-up vs
P⫽.8 Splint plus laser: from 30.83⫾7.73 at baseline to
30.49⫾6.93 at 3mo follow-up†
Ultrasound
Bakhtiary Ultrasound 15 treatment Low-level laser Pain (VAS) (4wk) P⬍.001, in favor Difference between ultrasound* and laser: ⫺4.4
and sessions in 3wk (n⫽45 therapy 15 of ultra-sound (95% CI, ⫺4.9 to ⫺3.1)
Rashidy- hands) treatment sessions
Pour38 in 3wk (n⫽45
hands)
Hand grip strength (4 P⬍.001 Difference between ultrasound* and laser: 12.1
weeks) (95% CI, 5.7 to 27.6)
Finger pinch (4wk) P⬍.001 Difference between ultrasound* and laser: 7.0 (95%
CI, 5.1 to 8.5)
Laser
Irvine Low-level laser therapy (3 Placebo laser Levine CTS P⫽.69 Unknown
et al39 times/wk for 5 wk; n⫽7) therapy 3 times/wk questionnaire
for 5 wk; n⫽8) Symptom severity
scale (4wk after
treatment)
Hand function (4wk Not significant Hand performance score: mean ⫾ SD, 81⫾9 to
after treatment) No P value 85⫾11 in treatment group vs 72⫾8 to 77⫾10 in
(Purdue pegboard test) given control group

Evcik Low-level laser therapy Placebo laser Pain (VAS) (12wk) Not significant No exact data given
et al40 (n⫽41; 5 times/wk for 2wk) therapy (n⫽40; 5
times/wk for 2wk)

Arch Phys Med Rehabil Vol 91, July 2010

1000 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede

APPENDIX 4: RECENT RCTs: CTS (Cont’d)

Outcome Measures Results:
Author Treatment Placebo Control/Comparison (Total Follow-Up Time) Statistical Results by Outcome Measure

Hand grip strength (kg) P⬎.05 Changes after 12wk: mean ⫾ SD, laser, 22.8⫾6.9,
(12wk) vs placebo, 19.6⫾7.3
Pinch grip (kg) (12wk) P⬎.05 Changes after 12wk: laser, 5.7⫾1.6, vs placebo:
4.8⫾1.5
Functional capacity Not significant No exact data given
(12wk)
Shooshtari Low-power laser therapy at Control group: flash Pain (VAS) (3wk) P⬍.001 Low-power laser: mean ⫾ SD, from 7.8⫾.42 before
et al41 anterior side of wrist and laser (n⫽40) treatment to 4.98⫾.12 after treatment(†)
palm for 15 sessions (5 P⬍.001 Control: from 8.01⫾.36 before treatment
times/wk; n⫽40)
to 7.62⫾0.4 after treatment(†)
Hand grip (kg) (3wk) P⬍.001 Low-power laser: mean ⫾ SD, from 19.81⫾5.06
before treatment to 22.86⫾5.13 after treatment(†)
P⫽.801 Control: from 21.46⫾6.23 before treatment to
21.52⫾6.05 after treatment(†)
Oral medication and
vitamins
Chang Prednisolone 20mg for 2 wk Prednisolone 20mg Overall improvement Not significant 49% improvement in 4-wk group vs 35.7%
et al48 and 10mg for 2 wk (n⫽53) for 2wk and placebo (12mo) improvement in 2-wk group at 12mo follow-up
Other nonsurgical
treatments
Bialosky NDT intended to provide Sham therapy Clinical pain (VAS) P⫽.73 NDT group: mean ⫾ SD, from 51.3⫾28.8 at
et al97 anatomical stress across (minimized baseline to 34.7⫾27.5 at 3wk follow-up vs sham
the median nerve 6 anatomical stress group: from 45.0⫾28.5 at baseline to 37.9⫾29.5 at
sessions in 3wk plus across the median 3wk follow-up
splinting (n⫽20) nerve) 6 sessions in NDT group: from 36⫾15.5 at baseline to 30.6⫾19.4
3wk plus splinting DASH Questionnaire P⬎.99 at 3wk follow-up vs sham group: from 41.3⫾19.0 at
(n⫽20) (3wk) baseline to 35.9⫾17.9 at 3wk follow-up
Grip strength P⫽.56 Mean grip strength: 22⫾9.6 at baseline to
24.9⫾10.0 at 3wk follow-up, no exact data given on
the 2 groups
Burke Graston Technique protocol Manual STM 2 Pain (VAS) (3mo after GISTM: mean ⫾ SD (95% CI), from
et al50 (GISTM), 2 treatments/wk treatments a week the last treatment 61.5⫾26.56 (46.5–76.5) at baseline to
for 4 wk followed by 1 for 4 weeks session) 9.2⫾11.04 (3.0–15.4) vs STM: from 60.5⫾17.9 (49.4–
treatment a week for 2 wk following 1 Range of motion (°) 71.6) at baseline to 33.7⫾ 28.84 (15.8–51.6) at 3mo†
and home exercises (n⫽12) treatment a week (3mo after the last Wrist extension
for 2 weeks and treatment session) GISTM: mean ⫾ SD (95% CI), from
home exercises 38.1⫾9.98 (32.5–43.7) at baseline to 43.9⫾.70 (37.8–
(n⫽10) 50.0) at 3mo vs STM: from 36.4⫾13.23 (28.2–44.6)
at baseline to 45.8⫾ 10.57 (39.2–52.4) at 3mo†
Wrist flexion
GISTM: mean ⫾ SD (95% CI), from
44.8⫾8.91 (39.3–49.8) at baseline to
49.9⫾8.44 (45.1–54.7) vs STM: from
47.5⫾8.05 (42.5–52.5) at baseline to 49.3⫾
12.2 (41.8–56.8) at 3mo†
Grip strength (3mo GISTM: mean ⫾ SD (95% CI), from
after the last treatment 20.2⫾8.79 (15.2–25.2) at baseline to
session) 25.4⫾7.67 (21.1–29.7) vs STM: from
23.5⫾6.78 (19.3–27.7) at baseline to 24.9⫾
6.32 (21.0–28.8) at 3mo†
Levine CTS GISTM: mean ⫾ SD (95% CI), from 2.1⫾.93 (1.6–
Questionnaire 2.6) at baseline to 1.6⫾.72 (1.2–2.0) at 3mo vs STM:
Functional scale (3mo from 2.4⫾.85 (1.9–2.9) at baseline to 1.7⫾.68 (1.3–
after the last treatment 2.1) at 3mo†
session)
Symptom severity GISTM: mean ⫾ SD (95% CI), from 3.0⫾.73 (2.6–
scale (3mo after the 3.4) at baseline to 1.8⫾.61 (1.5–2.1) at 3mo vs STM:
last treatment session) from 2.7⫾.64 (2.3–3.1) at baseline to 2.2⫾.59 (1.8–
2.6) at 3mo†
Weintraub Dynamic magnetic field Sham therapy Pain (VAS) (2mo) Not significant Treatment group: mean ⫾ SD, from 6.82⫾2.08 at
and stimulation to the wrist 4 (n⫽19) No P value baseline to 4.15⫾2.13 at 2mo vs sham therapy:
Cole55 h/d for 2mo (n⫽17) given from 5.17⫾1.54 at baseline to 3.78⫾2.27 at 2mo
Neuropathy Pain Scale P⫽.04 Reduction: treatment group* 42% vs controls 24%
Neuropathy Pain Scale
Total composite (2mo)
Patients Clinical Global P⫽.12 No exact data given
Impression of Change
(2mo)

Yang Acupuncture in 8 sessions 20mg prednisolone Global symptom score P⫽.15 Change (%) from baseline to 4wk: Acupuncture
et al57 over 4wk (n⫽38) daily for 2wk (4wk) group: mean ⫾ SD, –70⫾24.6, vs steroid group,
followed by 10 mg –64.7⫾27.6
prednisolone for
2wk (n⫽39)

Arch Phys Med Rehabil Vol 91, July 2010

 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
Author Treatment
Moraska TM protocol twice-weekly
et al58 30-min sessions for 6wk
(n⫽14)
Michlovitz Heat wrap with heat to
et al60 104°F (40°C) for 8 hours
daily (this temperature
maintained continuously
because of exposure to air)
for 3d (n⫽10)
Michalsen Cupping therapy (n⫽26)
et al61
Breuer Injection with botulinum
et al62 toxin B into each of the 3
hypothenar muscles groups
(n⫽11)
1.2500U botulinum B (n⫽9)
2.5000U botulinum B (n⫽1)
3.7500U botulinum B (n⫽1)
Amirjani Iontophoresis with 0.4%
et al64 dexamethasone sodium
phosphate with distilled
water (treatment group;
n⫽9)
Field 15-min massage once a
et al59 week for a 4-wk period and
self-massage daily (n⫽8)
1001
APPENDIX 4: RECENT RCTs: CTS (Cont’d)
Outcome Measures Results:
Placebo Control/Comparison (Total Follow-Up Time) Statistical Results by Outcome Measure
GM twice-weekly Grip strength (10wk) P⫽.04 TM group*: mean (95% CI), from 25.1 at baseline
30-min sessions for to 29.5kg (27.7–31.3 kg) at 12wk follow-up vs GM
6wk (n⫽13) group: from 25.1kg (mean) to 26.3kg at 12wk
follow-up
Pinch strength (6wk) P⫽.11 TM group: mean (95% CI), from 6.6kg at baseline
to 8.3kg (7.7–8.9) at 6wk follow-up vs GM group:
from 6.6kg at baseline to 7.2kg (6.6–7.8) at 6wk
follow-up
Levine CTS P⫽.80 TM group: mean (95% CI), from 2.3 at baseline to
Questionnaire 1.8 (1.6–2.0) at 10wk follow-up vs GM group: from
Symptom severity 2.3 at baseline to 1.9 (1.7–2.1) at 10wk follow-up
score (10wk)
Function status scale P⫽.34 Only the following data are given: TM group: mean
(10wk) ⫾ SD, 1.6⫾0.2 after 4wk follow-up. GM group:
1.7⫾0.2 after 6wk follow-up
Grooved Pegboard Test P⫽.41 TM group: mean ⫾ SD, from 104 at baseline to
(10wk) 98⫾4.8 after 6wk follow-up vs GM group: from 104
at baseline to 95⫾5.2 after 6wk follow-up
Oral placebo for 3d Pain relief (5d) P⫽.001 Heat wrap*: mean ⫾ SD, 2.18⫾.34 vs oral placebo:
4 times daily, 2 0.95⫾.25 at day 1 to 3/hour 0 to 8 daily (this
tablets (n⫽12) temperature maintained
Pⱕ.05 Heat wrap* vs oral placebo at 20 of 26 time points
Joint stiffness P⫽.004 Heat wrap*: mean ⫾ SD, 21.8⫾5.5 vs oral placebo:
reduction (5d) 4.9⫾3.1 at day 1 to 3/hour 0 to 8
Pⱕ.05 Heat wrap* vs oral placebo at 19 of 26 time points
Grip strength (5d) P⫽.012 Heat wrap*: mean ⫾ SD, 6.1⫾1.6kg vs oral
placebo: 0.8⫾1.4kg at 5d follow-up
Patient-rated wrist P⫽.013 Heat wrap*: mean ⫾ SD, 27.3⫾5.9 vs oral placebo:
evaluation 7.9⫾ 5.39 at 5d follow-up
Levine CTS Pⱕ.001 Heat wrap*: mean ⫾ SD, .97⫾.16 vs oral placebo:
Questionnaire .14⫾.14 at 5d follow-up
Symptom severity
scale
Function status scale P⫽.006 Heat wrap: mean ⫾ SD, .65⫾.16 vs oral placebo:
.00⫾.16 at 3d follow-up
P⫽.07 Heat wrap: .57⫾.22 vs oral placebo: .12⫾.20 at day
5
Heating pad (control Pain at rest (7d) Significant Cupping* vs heating pad: differences at day 7:
Group; n⫽26) MD⫽⫺22.9 (95% CI, ⫺35.3 to ⫺10.5)
Levine CTS Significant Cupping* vs heating pad: differences at day 7:
Questionnaire MD⫽⫺0.6 (95% CI, ⫺0.9 to ⫺0.2)
Symptom severity (7d)
Functional status Significant Cupping* vs heating pad: differences at day 7:
MD⫽⫺0.6 (95% CI, ⫺0.8 to ⫺0.3)
DASH score Significant Cupping* vs heating pad: differences at day 7:
MD⫽⫺11.1 (95% CI, ⫺17.1 to ⫺5.1)
(n⫽9) West Haven-Yale Not significant No exact data given
Multidimensional Pain No P value
Inventory (13wk) given
Iontophoresis with Levine CTS P⫽.73 Treatment group: mean (25%–75% CI), 38 (31–40)
distilled water Questionnaire (6mo) vs control group: 36 (33–54) at baseline
(control group; n⫽8)
P⫽.97 Treatment group: 33 (24–44) vs control group:
32 (26–37) at 3mo follow-up
P⫽.25 Treatment group: 26 (24–31) vs control group:
34 (22–41) at 6mo follow-up
(n⫽8) Pain (VITAS) (at 4wk P⬍.05 Massage group*: from 4.11 at baseline to 2.59 at
follow-up after 3 4wk follow-up vs controls: from 6.17 at baseline to
treatment sessions) 4.83 at 4wk follow-up
Carpal tunnel P⬍.05 Massage group*: from 3.00 at the first day to 2.22
symptoms (4wk) at the last day vs controls: from 3.00 at the first
day to 3.00 at the last day
Grip strength (at 4wk P⬍.05 Massage group*: from 6.61 at baseline to 7.8 at
follow-up after (at 4wk 4wk follow-up vs controls: 5.58 at baseline to 6.25
follow-up after at 4wk follow-up
Arch Phys Med Rehabil Vol 91, July 2010
1002 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede

APPENDIX 4: RECENT RCTs: CTS (Cont’d)

Outcome Measures Results:
Author Treatment Placebo Control/Comparison (Total Follow-Up Time) Statistical Results by Outcome Measure

Corticosteroid injections
Hui et al69 local injection of 15mg
methylprednisolone into the
carpal tunnel with oral
placebo for 10d (n⫽30)
Long-term results from the
study of Wong et al68
Dammers 3 doses of
et al73 methylprednisone
compared: 20mg vs 40mg
vs 60mg (n⫽45, n⫽43,
n⫽44, respectively)
Local injection with Global symptom score P⫽.07 Steroid injection: mean ⫾ SD, 25.0⫾6.41 at aseline
normal saline with (80wk) to 13.57⫾7.47 at 2wk follow-up vs oral
oral prednisolone prednisolone: 25.73⫾8.31 at baseline to 17.77⫾
25mg daily for 10d 9.98 at 2wk follow-up.
(n⫽30)
P⫽.004 Steroid injection: mean ⫾ SD, 14.30⫾8.42 at 12 wk†
P⬎.05 Steroid injection: 69.5% improvement compared
with baseline vs oral prednisolone: 51.9%
improvement compared with baseline after 7 mo in
participants who did not require surgery (n⫽19 and
n⫽16, respectively)
Free of symptoms (1y) P⬍.05 6mo: 60mg group* 73% (32/44) vs 40mg group
53% (23/43)

Treatment response P⬍.05 60mg group* 73% (32/44) vs 20mg group 56% (25/
symptoms (1y) P⫽.4711 45) 1y: 20mg group 47% vs 40mg group 41% vs
60mg group 52%
Moghtaderi One injection of 40mg Daily application of Pain (VAS) (30d) P⬍.001 EMLA: mean ⫾ SD, from 5.8⫾.98 at baseline to
et al79 Methylprednisolone acetate EMLA cream on the P⬍.001 2.1⫾1.2 at 4wk follow-up vs Injection: mean ⫾ SD,
at the wrist (n⫽35) volar aspect of the from 5.7⫾1.0 at baseline to 1.6⫾1.4 at 4wk follow-
wrist up†

Abbreviations: d, days; EMLA, Eutectic mixture of Local Anesthetic; GM, general massage protocol; GISTM, Graston instrument–assisted soft tissue mobilization; MCP,
metacarpophalangeal; mo, month; NDT, neurodynamic technique; STM, soft tissue mobilization; TM, targeted massage; VITAS, “Lives” innovative hospital care; wk, weeks.
*In favor of.
†
No comparison between the groups was made.

References

1. Huisstede BM, Miedema HS, Verhagen AP, Koes BW, Verhaar
JA. Multidisciplinary consensus on the terminology and classifi-
cation of complaints of the arm, neck and/or shoulder. Occup
Environ Med 2007;64:313-9.
2. Huisstede BM, Wijnhoven HA, Bierma-Zeinstra SM, Koes BW,
Verhaar JA, Picavet S. Prevalence and characteristics of com-
plaints of the arm, neck, and/or shoulder (CANS) in the open
population. Clin J Pain 2008;24:253-9.
3. Shiri R, Miranda H, Heliovaara M, Viikari-Juntura E. Physical
work load factors and carpal tunnel syndrome: a population-based
study. Occup Environ Med 2009;66:368-73.
4. Palmer DH, Hanrahan LP. Social and economic costs of carpal
tunnel surgery. Instr Course Lect 1995;44:167-72.
5. Phalen S. The carpal tunnel syndrome: seventeen years’ experi-
ence in diagnosis and treatment of six hundred fifty-four hands.
J Bone Joint Surg Am 1966:211-28.
6. Werner RA, Andary M. Carpal tunnel syndrome: pathophysiology
and clinical neurophysiology. Clin Neurophysiol 2002;113:1373-
81.
7. van Rijn RM, Huisstede BM, Koes BW, Burdorf A. Associations
between work-related factors and the carpal tunnel syndrome—a
systematic review. Scand J Work Environ Health 2009;35:19-36.
8. Rosenbaum RB, Ochoa JL. Carpal tunnel syndrome and other
disorders of the median nerve. Stoneham: Butterworth-
Heinemann; 1993.
9. Practice parameter for carpal tunnel syndrome (summary state-
ment). Report of the Quality Standards Subcommittee of the
American Academy of Neurology. Neurology 1993;43:2406 –9.
10. de Krom MC, van Croonenborg JJ, Blaauw G, Scholten RJ,
Spaans F. [Guideline “Diagnosis and treatment of carpal tunnel
syndrome”]. Ned Tijdschr Geneeskd 2008;152:76-81.
11. Bland JD. Carpal tunnel syndrome. BMJ 2007;335:343-6.
12. Miller RS, Iverson DC, Fried RA, Green LA, Nutting PA. Carpal
tunnel syndrome in primary care: a report from ASPN. Ambula-
tory Sentinel Practice Network. J Fam Pract 1994;38:337-44.
13. O’Connor D, Marshall S, Massy-Westropp N. Non-surgical treat-
ment (other than steroid injection) for carpal tunnel syndrome.
Cochrane Database Syst Rev 2003(1):CD003219.
14. Marshall S, Tardif G, Ashworth N. Local corticosteroid injection
for carpal tunnel syndrome. Cochrane Database Syst Rev 2007(2):
CD001554.
15. Furlan AD, Pennick V, Bombardier C, van Tulder M. 2009
Updated method guidelines for systematic reviews in the Co-
chrane Back Review Group. Spine (Phila Pa 1976) 2009;34:1929-
41.
16. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated
method guidelines for systematic reviews in the Cochrane Col-
laboration Back Review Group. Spine 2003;28:1290-9.
17. Piravej K, Boonhong J. Effect of ultrasound thermotherapy in
mild to moderate carpal tunnel syndrome. J Med Assoc Thai
2004;87(Suppl 2):S100-6.
18. Arinci Incel N, Sezgin M, Aksit C, Sahin G. Effect of gabapentin
in carpal tunnel syndrome: a controlled clinical trial. J Rheumatol
Med Rehabil 2005:224-30.
19. Avci S, Gunaydin R, Oztura I. Comparison of effectiveness of
splint and splint with physical therapy in carpal tunnel syndrome.
FTR - Turkiye Fiziksel Tip ve Rehabilitasyon Dergisi 2004:22-6.
20. Clarke M, Osman A. Cochrane reviewers’ handbook 4.0 [updated
July 1999]. In: The Cochrane library, the Cochrane collaboration.
Oxford: Update Software; 1999(Issue 2).
21. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of
reports of randomized clinical trials: is blinding necessary? Con-
trol Clin Trials 1996;17:1-12.
22. Burke DT, Burke MM, Stewart GW, Cambre A. Splinting for
carpal tunnel syndrome: in search of the optimal angle. Arch Phys
Med Rehabil 1994;75:1241-4.
23. Manente G, Torrieri F, Di Blasio F, Staniscia T, Romano F,
Uncini A. An innovative hand brace for carpal tunnel syndrome:
a randomized controlled trial. Muscle Nerve 2001;24:1020-5.

Arch Phys Med Rehabil Vol 91, July 2010

 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
24. Walker WC, Metzler M, Cifu DX, Swartz Z. Neutral wrist splint-
ing in carpal tunnel syndrome: a comparison of night-only versus
full-time wear instructions. Arch Phys Med Rehabil 2000;81:
424-9.
25. Baysal O, Altay Z, Ozcan C, Ertem K, Yologlu S, Kayhan A.
Comparison of three non-surgical treatment protocols in carpal
tunnel syndrome. Int J Clin Pract 2006;60:820-8.
26. Brininger TL, Rogers JC, Holm MB, Baker NA, Li ZM, Goitz RJ.
Efficacy of a fabricated customized splint and tendon and nerve
gliding exercises for the treatment of carpal tunnel syndrome: a
randomized controlled trial. Arch Phys Med Rehabil 2007;88:
1429-35.
27. Heebner ML, Roddey TS. The effects of neural mobilization in
addition to standard care in persons with carpal tunnel syndrome
from a community hospital. J Hand Ther 2008;21:229-40; quiz 41.
28. Pinar L, Enhos A, Ada S, Gungor N. Can we use nerve gliding
exercises in women with carpal tunnel syndrome? Adv Ther
2005;22:467-75.
29. Premoselli S, Sioli P, Grossi A, Cerri C. Neutral wrist splinting in
carpal tunnel syndrome: a 3- and 6-months clinical and neuro-
physiologic follow-up evaluation of night-only splint therapy.
Eura Medicophys 2006;42:121-6.
30. Yagci I, Elmas O, Akcan E, Ustun I, Gunduz OH, Guven Z.
Comparison of splinting and splinting plus low-level laser therapy
in idiopathic carpal tunnel syndrome. Clin Rheumatol 2009;28:
1059-65.
31. De Angelis MV, Pierfelice F, Di Giovanni P, Staniscia T, Uncini
A. Efficacy of a soft hand brace and a wrist splint for carpal tunnel
syndrome: a randomized controlled study. Acta Neurol Scand
2009;119:68-74.
32. Akalin E, El O, Peker O, et al. Treatment of carpal tunnel
syndrome with nerve and tendon gliding exercises. Am J Phys
Med Rehabil 2002;81:108-13.
33. Mishra S, Prabhakar S, Lal V, Modi M, Das CP, Khurana D.
Efficacy of splinting and oral steroids in the treatment of carpal
tunnel syndrome: a prospective randomized clinical and electro-
physiological study. Neurol India 2006;54:286-90.
34. Garfinkel MS, Singhal A, Katz WA, Allan DA, Reshetar R,
Schumacher HR, Jr.Yoga-based intervention for carpal tunnel
syndrome: a randomized trial. JAMA 1998;280:1601-3.
35. Ebenbichler GR, Resch KL, Nicolakis P, et al. Ultrasound treat-
ment for treating the carpal tunnel syndrome: randomised “sham”
controlled trial. BMJ 1998;316:731-5.
36. Oztas O, Turan B, Bora I, Karakaya MK. Ultrasound therapy
effect in carpal tunnel syndrome. Arch Phys Med Rehabil 1998;
79:1540-4.
37. Koyuncu HU, FN. Sahin, U. Togay, P. 1Mhz - 3MHz ultrasound
applications in carpal tunnel syndrome [Karpal tunel sendromunda
1MHz - 3MHz ultrason uygulamasi]. Fizik Tedavi ve Rehabilita-
syon Dergisi 1995;19:141–5.
38. Bakhtiary AH, Rashidy-Pour A. Ultrasound and laser therapy in
the treatment of carpal tunnel syndrome. Aust J Physiother 2004;
50:147-51.
39. Irvine J, Chong SL, Amirjani N, Chan KM. Double-blind ran-
domized controlled trial of low-level laser therapy in carpal tunnel
syndrome. Muscle Nerve 2004;30:182-7.
40. Evcik D, Kavuncu V, Cakir T, Subasi V, Yaman M. Laser therapy
in the treatment of carpal tunnel syndrome: a randomized con-
trolled trial. Photomed Laser Surg 2007;25:34-9.
41. Shooshtari SM, Badiee V, Taghizadeh SH, Nematollahi AH,
Amanollahi AH, Grami MT. The effects of low level laser in
clinical outcome and neurophysiological results of carpal tunnel
syndrome. Electromyogr Clin Neurophysiol 2008;48:229-31.
42. Chang MH, Chiang HT, Lee SS, Ger LP, Lo YK. Oral drug of
choice in carpal tunnel syndrome. Neurology 1998;51:390-3.
1003
43. Herskovitz S, Berger AR, Lipton RB. Low-dose, short-term oral
prednisone in the treatment of carpal tunnel syndrome. Neurology
1995;45:1923-5.
44. Hui AC, Wong SM, Wong KS, et al. Oral steroid in the treatment
of carpal tunnel syndrome. Ann Rheum Dis 2001;60:813-4.
45. Pal BM, P. Hossain, MA. Wallace, AS. Diffey, BL. Should
diuretics be prescribed for idiopathic carpal tunnel syndrome?
results of a controlled trial. Clin Rehabil 1988;2:299 –301.
46. Spooner GR, Desai HB, Angel JF, Reeder BA, Donat JR. Using
pyridoxine to treat carpal tunnel syndrome: randomized control
trial. Can Fam Physician 1993;39:2122-7.
47. Stransky M, Rubin A, Lava NS, Lazaro RP. Treatment of carpal
tunnel syndrome with vitamin B6: a double-blind study. South
Med J 1989;82:841-2.
48. Chang MH, Ger LP, Hsieh PF, Huang SY. A randomised clinical
trial of oral steroids in the treatment of carpal tunnel syndrome: a
long term follow up. J Neurol Neurosurg Psychiatry 2002;73:
710-4.
49. Tal-Akabi A, Rushton A. An investigation to compare the effec-
tiveness of carpal bone mobilisation and neurodynamic mobilisa-
tion as methods of treatment for carpal tunnel syndrome. Man
Ther 2000;5:214-22.
50. Burke DJ, Buchberger DC, Carey-Loghmani MT, Dougherty PE,
Douglas SG, Dishman JD. A pilot study comparing two manual
therapy interventions for carpal tunnel syndrome. J Manipulative
Physiol Ther 2007;30:50-61.
51. Davis PT, Hulbert JR, Kassak KM, Meyer JJ. Comparative effi-
cacy of non-surgical medical and chiropractic treatments for car-
pal tunnel syndrome: a randomized clinical trial. J Manipulative
Physiol Ther 1998;21:317-26.
52. Rempel D, Tittiranonda P, Burastero S, Hudes M, So Y. Effect of
keyboard keyswitch design on hand pain. J Occup Environ Med
1999;41:111-9.
53. Tittiranonda P, Rempel D, Armstrong T, Burastero S. Effect of
four computer keyboards in computer users with upper extremity
musculoskeletal disorders. Am J Ind Med 1999;35:647-61.
54. Carter R, Aspy CB, Mold J. The effectiveness of magnet therapy
for treatment of wrist pain attributed to carpal tunnel syndrome. J
Fam Pract 2002;51:38-40.
55. Weintraub MI, Cole SP. A randomized controlled trial of the
effects of a combination of static and dynamic magnetic fields on
carpal tunnel syndrome. Pain Med 2008;9:493-504.
56. Aigner NZ, G. Petje, G. Results of laser-acupuncture in carpal
tunnel syndrome: a prospective, randomised and blinded study
[Laserakupunktur bei der praoperativen schmerzbekampfung
beim karpaltunnelsyndrom: eine prospektiv randomisierte studie].
Deutsche Zeitschrift für Akupunktur 1999;42:70 –5.
57. Yang CP, Hsieh CL, Wang NH, et al. Acupuncture in patients
with carpal tunnel syndrome: a randomized controlled trial. Clin J
Pain 2009;25:327-33.
58. Moraska A, Chandler C, Edmiston-Schaetzel A, Franklin G,
Calenda EL, Enebo B. Comparison of a targeted and general
massage protocol on strength, function, and symptoms associated
with carpal tunnel syndrome: a randomized pilot study. J Altern
Complement Med 2008;14:259-67.
59. Field T, Diego M, Cullen C, et al. Carpal tunnel syndrome
symptoms are lessened following massage therapy. J Bodyw Mov
Ther 2004;8:9-14.
60. Michlovitz S, Hun L, Erasala GN, Hengehold DA, Weingand
KW. Continuous low-level heat wrap therapy is effective for
treating wrist pain. Arch Phys Med Rehabil 2004;85:1409-16.
61. Michalsen A, Bock S, Ludtke R, et al. Effects of traditional
cupping therapy in patients with carpal tunnel syndrome: a ran-
domized controlled trial. J Pain 2009;10:601-8.
62. Breuer B, Sperber K, Wallenstein S, et al. Clinically significant
placebo analgesic response in a pilot trial of botulinum B in
Arch Phys Med Rehabil Vol 91, July 2010
1004 EFFECTIVENESS OF NONSURGICAL TREATMENTS FOR CARPAL TUNNEL SYNDROME, Huisstede
patients with hand pain and carpal tunnel syndrome. Pain Med
2006;7:16-24.
63. Ozkul Y, Sabuncu T, Yazgan P, Nazligul Y. Local insulin injec-
tion improves median nerve regeneration in NIDDM patients with
carpal tunnel syndrome. Eur J Neurol 2001;8:329-34.
64. Amirjani N, Ashworth NL, Watt MJ, Gordon T, Chan KM.
Corticosteroid iontophoresis to treat carpal tunnel syndrome: a
double-blind randomized controlled trial. Muscle Nerve 2009;39:
627-33.
65. Dammers JW, Veering MM, Vermeulen M. Injection with meth-
ylprednisolone proximal to the carpal tunnel: randomised double
blind trial. BMJ 1999;319:884-6.
66. Armstrong T, Devor W, Borschel L, Contreras R. Intracarpal
steroid injection is safe and effective for short-term management
of carpal tunnel syndrome. Muscle Nerve 2004;29:82-8.
67. Ozdogan H, Yazici H. The efficacy of local steroid injections in
idiopathic carpal tunnel syndrome: a double-blind study. Br J
Rheumatol 1984;23:272-5.
68. Wong SM, Hui AC, Tang A, et al. Local vs systemic corticoste-
roids in the treatment of carpal tunnel syndrome. Neurology
2001;56:1565-7.
69. Hui AC, Wong SM, Tang A, Mok V, Hung LK, Wong KS.
Long-term outcome of carpal tunnel syndrome after non-surgical
treatment. Int J Clin Pract 2004;58:337-9.
70. Celiker R, Arslan S, Inanici F. Corticosteroid injection vs. non-
steroidal antiinflammatory drug and splinting in carpal tunnel
syndrome. Am J Phys Med Rehabil 2002;81:182-6.
71. Lucantoni CG, S. Gaetti, R. Comparison between He-Ne laser
therapy and steroid injections in the treatment of idiopathic carpal
tunnel syndrome [Confronto tra laserterapia HE-NE infiltrativa
steroidea nel trattamento della sindrome idiopatica del tunnel
carpale]. La Riabilitazione 1992;25:249 –56.
72. O’Gradaigh D, Merry P. Corticosteroid injection for the treatment
of carpal tunnel syndrome. Ann Rheum Dis 2000;59:918-9.
73. Dammers JW, Roos Y, Veering MM, Vermeulen M. Injection
with methylprednisolone in patients with the carpal tunnel syn-
drome: a randomised double blind trial testing three different
doses. J Neurol 2006;253:574-7.
74. Wong SM, Hui AC, Lo SK, Chiu JH, Poon WF, Wong L. Single
vs. two steroid injections for carpal tunnel syndrome: a random-
ised clinical trial. Int J Clin Pract 2005;59:1417-21.
75. Sevim S, Dogu O, Camdeviren H, et al. Long-term effectiveness
of steroid injections and splinting in mild and moderate carpal
tunnel syndrome. Neurol Sci 2004;25:48-52.
76. Habib GS, Badarny S, Rawashdeh H. A novel approach of local
corticosteroid injection for the treatment of carpal tunnel syn-
drome. Clin Rheumatol 2006;25:338-40.
77. Aygul R, Ulvi H, Karatay S, Deniz O, Varoglu AO. Determination
of sensitive electrophysiologic parameters at follow-up of differ-
ent steroid treatments of carpal tunnel syndrome. J Clin Neuro-
physiol 2005;22:222-30.
78. Gökoğlu F, Fndkoglu G, Yorgancoglu ZR, Okumus M, Ceceli E,
Kocaoglu S. Evaluation of iontophoresis and local corticosteroid
injection in the treatment of carpal tunnel syndrome. Am J Phys
Med Rehabil 2005;84:92-6.
79. Moghtaderi AR, Jazayeri SM, Azizi S. EMLA cream for carpal
tunnel syndrome: how it compares with steroid injection. Electro-
myogr Clin Neurophysiol 2009;49:287-9.
80. van Middelkoop M, Huisstede BM, Glerum S, Koes BW. Effec-
tiveness of interventions of specific complaints of the arm, neck,
Arch Phys Med Rehabil Vol 91, July 2010
or shoulder (CANS): musculoskeletal disorders of the hand. Clin
J Pain 2009;25:537-52.
81. Huisstede BM, van Middelkoop M, Randsdorp M, Glerum S,
Koes BW. Effectiveness of interventions of specific complaints of
the arm, neck and/or shoulder: 3 musculoskeletal disorders of the
hand. An update. Arch Phys Med Rehabil. 2010 Feb;91:298-314.
82. Bisset L, Beller E, Jull G, Brooks P, Darnell R, Vicenzino B.
Mobilisation with movement and exercise, corticosteroid injec-
tion, or wait and see for tennis elbow: randomised trial. BMJ
Clinical research ed. 2006;333:939.
83. Smidt N, van der Windt DA, Assendelft WJ, Deville WL,
Korthals-de Bos IB, Bouter LM. Corticosteroid injections, phys-
iotherapy, or a wait-and-see policy for lateral epicondylitis: a
randomised controlled trial. Lancet 2002;359:657-62.
84. Nichols AW. Complications associated with the use of corticoste-
roids in the treatment of athletic injuries. Clin J Sport Med
2005;15:370-5.
85. Watson T. Ultrasound in contemporary physiotherapy practice.
Ultrasonics 2008;48:321-9.
86. Baldes EJ, Herrick JF, Stroebel CF, 3rd.Biologic effects of ultra-
sound. Am J Phys Med 1958;37:111-21.
87. Lehmann J, de Lateur B. Therapeutic heat. In: Lehmann JF, editor.
Therapeutic heat and cold. 2nd ed. Baltimore: Williams &
Wilkins; 1982. p 386-403.
88. Dunn F, Schwan HP. A quantitative similarity between some
biological effects of ultrasound and microwaves. Br J Cancer
1982;5:93-4.
89. Lubinska L, Olekiewicz M. Rate of regeneration of amphibian
peripheral nerves at different temperatures. Acta Biol Exp 1950;
15:125-45.
90. Hong CZ. Reversible nerve conduction block in patients with
polyneuropathy after ultrasound thermotherapy at therapeutic dos-
age. Arch Phys Med Rehabil 1991;72:132-7.
91. Kruger VL, Kraft GH, Deitz JC, Ameis A, Polissar L. Carpal
tunnel syndrome: objective measures and splint use. Arch Phys
Med Rehabil 1991;72:517-20.
92. Pascarelli EF, Hsu YP. Understanding work-related upper extrem-
ity disorders: clinical findings in 485 computer users, musicians,
and others. J Occup Rehabil 2001;11:1-21.
93. Boutron I, Moher D, Tugwell P, et al. A checklist to evaluate a
report of a nonpharmacological trial (CLEAR NPT) was devel-
oped using consensus. J Clin Epidemiol 2005;58:1233-40.
94. Verhagen AP, de Vet HC, de Bie RA, Kessels AG, Boers M,
Knipschild PG. Balneotherapy and quality assessment: interob-
server reliability of the Maastricht criteria list and the need for
blinded quality assessment. J Clin Epidemiol 1998;51:335-41.
95. van Tulder MW, Suttorp M, Morton S, Bouter LM, Shekelle P.
Empirical evidence of an association between internal validity and
effect size in randomized controlled trials of low-back pain. Spine
(Phila Pa 1976) 2009;34:1685-92.
96. Furlan AD, Pennick V, Bombardier C, van Tulder M, Editorial
Board, Cochrane Back Review Group. 2009 updated method
guidelines for systematic reviews in the Cochrane Back Review
Group. Spine (Phila Pa 1976). 2009 Aug 15;34:1929-41.
97. Bialosky JE, Bishop MD, Price DD, Robinson ME, Vincent KR,
George SZ. A randomized sham-controlled trial of a neurody-
namic technique in the treatment of carpal tunnel syndrome.
J Orthop Sports Phys Ther 2009;39:709-23.