540
C. Minns Lowe et al.
genetic (15) or endocrine disorders, such as diabetes
and thyroid problems that may become more prevalent
at this time (11). FS has been described to occur in 3
transitionary phases: increasing pain and progressive
stiffness, ongoing stiffness and decreasing pain, and a
resolution phase, in which the remaining pain settles,
and movement improves (2, 8, 16). Lewis (17) sug-
gested a simplified method of classification; a pain
greater than stiffness phase and a stiffness greater
than pain phase. When there is no known reason for
onset, such as following surgery, or if the condition is
not associated with comorbidities, the term primary
(or alternatively idiopathic) FS is recommended (18).
Whether FS is a self-limiting condition that may do
well without intervention (7) or whether resolution re-
quires treatment remains unclear (19). A retrospective
study suggests that a “no treatment” option may be con-
sidered (20): 94% of people (n = 83, mean time after
onset = 9 years) with FS, recovered to normal levels of
function and motion without treatment. Management
options for FS broadly fall into 3 categories; advice,
support and empathy, while allowing natural history
to take its course (wait and watch); more formal non-
surgical management; and surgical intervention.
A comprehensive, high-quality systematic review
of non-surgical management for people with FS con-
cluded that data from studies with a low risk of bias
were sparse (21). The following conclusions were
drawn from a minority of studies deemed to be of low
risk of bias (21): possible short-term benefit from ad-
ding a single intra-articular steroid injection to home
exercise for patients with primary FS of less than 6
months’ duration, adding physiotherapy (including
mobilization in 8–10 sessions over 4 weeks) to a single
steroid injection, adding shortwave diathermy (SWD)
to passive mobilization and home exercise (for some
outcomes only), and high-grade mobilization may be
more effective than low-grade mobilization in a popu-
lation of patients who have already had physiotherapy
and/or steroid injection (for some outcomes).
Recommendations to classify FS into diagnostic sub-
categories have been made, with the aim of developing
more homogeneous research investigations and better
understanding of whether differences in pathogenesis
and management exist within the different subgroups
(17). The term primary or idiopathic FS appears to be the
most common presentation and is the diagnostic category
when there is no identifiable reason for onset. Secondary
FS is used when there is an identifiable potential cause
preceding the onset of the condition. Secondary systemic
FS describes FS in the presence of diabetes. When the
condition is preceded by a humeral or clavicular fracture,
a chest wall tumour, cervical radiculopathy, ipsilateral
breast surgery or cerebrovascular accident, the diagnostic
www.medicaljournals.se/jrm
label secondary extrinsic FS has been suggested. FS oc-
curring post-surgery is termed iatrogenic FS.
The current review aims to update the existing
review (21) and evaluate the clinical effectiveness of
non-surgical management interventions focussing on
primary or idiopathic FS, in terms of pain, range of
shoulder joint movement, self-reported function and
disability, quality of life, recovery time, return to work
and recreation and adverse events.
METHODS
This review was carried out following recommended advice
from the Cochrane Handbook, reported according to the PRIS-
MA statement (22) and registered in advance with PROSPERO
(reference number: CRD42015013728).
Population
Randomized controlled trials (RCTs) containing participants
with primary (idiopathic) FS or its synonyms, such as adhesive
capsulitis, were included. FS in people with diabetes has been
considered as both primary and secondary. For this review,
unlike Maund et al. (21), but in agreement with a consensus
on defining the subcategories of FS (18), people with diabetes
were considered to have systemic secondary FS and these
studies were excluded. Participants with other non-idiopathic
or secondary causes of FS (for example trauma or surgery), as
well as participants with symptoms indicating an alternative
source of shoulder pain (for example, referred pain, paraesthesia,
instability, rotator cuff tendinopathy) were excluded. Studies
with participants with general shoulder symptoms (shoulder
pain or symptoms or non-specific shoulder pain) were not in-
cluded. Studies needed to state they were RCTs; studies stating
participants were “divided”’ or “assigned”’ to groups were not
assumed to be RCTs and were not included.
Non-surgical interventions
Physiotherapy and rehabilitative interventions, distension, oral
medications, and injection procedures were included. Surgical
procedures, such as arthroscopic and open capsular releases,
were excluded.
Comparators
Trials comparing an intervention and control group, or, com-
paring 2 of the included interventions were included. Trials
comparing an intervention, a further intervention outside of the
review and a control were included only if the data could be
extracted for the intervention and control arms.
Outcomes
Outcomes were: adverse events, pain, range of shoulder joint
motion, self-reported function and disability, strength, quality
of life, recovery time to recreation and return to work.
Data sources and search strategy
Electronic searches were initially conducted in December 2014
by 2 independent reviewers (EB, NDB) and subsequently upda-
ted by 2 independent reviewers between November 2016 and