C. Minns Lowe et al.
cycle ergometer (n = 29), compared
with ultrasound, self-assisted exercises,
Codman exercise, Swiss ball exercises
and isometric exercises (n = 28) (Table
I). In this study, the participants were po-
Records excluded
(n=21)
sitioned in supine in 30–40° of shoulder
Not an RCT n=5
abduction and external rotation (accor-
Duplicate n=9
Records screened for
Not published in
ding to tolerance). A Kaltenborn type III
conservative review
English:
axial distraction was applied, followed
(n=76)
Chinese n=5
French n=1
by a posterior glide, without oscillations,
German n=1
for 1 min. This was repeated 15 times,
with a 1-min rest, and 10 sessions were
delivered, 2 or 3 times per week. Pain,
Full-text articles
assessed for eligibility
range of motion, and function improved
(n=55)
with statistical differences in favour of
the mobilization and cycle ergometer
group (Table I). There was no loss to
follow-up.
Trials included in review n=30
Full-text articles excluded (n=23)
Only the short-term results were re-
Unable to clarify/separate diabetic &
ported (at the end of the 10 th treatment
Trial follow up papers n=2
non- diabetic data n=16
Unable to separate primary FS from
session). The authors reported a mean
Studies evaluated as low risk of bias
secondary/other shoulder conditions n=3
increase, in favour of the mobilization
and included in synthesis (n= 4)
Excluded adhesive capsulitis n=1
Celik & Kaya Mutlu (28)
Unable to separate trauma data n=1
and ergometer group for the primary
Gutierrez Espinosa et al. (15)
Unable to separate trauma and diabetes
Ibrahim et al. (31)
data n=1
outcome measures; passive external
Yoon et al. (32)
Unclear if RCT n=1
rotation of 27° (56.8° vs 30°), passive
flexion of 37° (107° vs 69.7°), passive
abduction of 22° (70.7° vs 48.8°), and
Fig. 1. PRISMA 2009 flow diagram. RCT: randomized controlled trial; FS: frozen shoulder.
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred
for the secondary outcome measure-
Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS
ments; VAS (pain) of 4.4 (0 = no pain
Med 6 (7): e1000097. doi:10.1371/journal.pmed1000097 (22).
and 10 = worst imaginable pain) in the
mobilization group and 5.4 in the ultra-
improved over time. The sample size (n = 30) was
sound and exercise group, and, 21 points
small and between-group comparisons showed that the
in the Constant-Murley Score (50.3 vs 29.7).
effect size for the primary outcome used in the power
Although the findings suggest clinical improve-
calculation, the Disabilities of the Arm, Shoulder and
ment in all the outcomes of interest in favour of the
Hand (DASH), was 0.03. Although the trend (8 points
mobilization and ergometer group, the absence of
at the end of treatment and 6 points at 12 months) in
medium- and long-term follow-up is a clear limita-
the DASH was better in the mobilization and stretching
tion of this study.
group over the stretching alone group this would not
be considered clinically meaningful. The mobiliza-
Comparison of static progressive stretching plus
tions and stretching group had improved Constant
multi-modal intervention vs multi-modal intervention
score and abduction and external rotation range at the
Ibrahim et al. (31) compared the effectiveness of a
end of treatment, and this was maintained at 1-year
multi-modal treatment programme consisting of heat
follow-up. However, it is unclear if the 14° increase
packs, therapy to facilitate muscle relaxation and gle-
in abduction, and 6° increase in external rotation at 1
nohumeral mobilizations (inferior glides, longitudinal
year is of clinical relevance, although the 17.2 increase
caudad) for 2 min, using large-amplitude oscillations
in Constant score might be relevant.
and repeated 3 times in a 10-min session (n = 30) and
a static progressive stretch device with the multi-modal
Comparison of joint mobilization and upper
programme alone (n = 30). Both groups received
extremity cycle ergometer vs ultrasound and
treatment 3 times a week for 4 weeks and participants
exercises
were also provided with a home exercise programme.
Gutierrez Espinosa et al. (30) also evaluated joint mo-
The static stretch group were asked to apply the static
bilization combined with 15 min on an upper extremity
stretch device for a single 30-min session daily for
Records identified through
MEDLINE database searching
until 2016
(n=2,724)
542
www.medicaljournals.se/jrm
Additional records identified
through other sources until 2017
(n=0)