Jump to content

Frozen Shoulder

Definition/Description

MRI: thickened joint capsule, especially at the inferior recess, can be a sign of frozen shoulder.

Frozen shoulder, also called adhesive capsulitis, is a condition in which painful, and later progressively restricted, active and passive range of motion of the glenohumeral (GH) joint develops, usually with external rotation most affected. Terminology is contested (adhesive capsulitis, painful stiff shoulder, periarthritis and idiopathic restriction of shoulder movement are all used) and there are no universally accepted diagnostic criteria.[1][2] This page uses "frozen shoulder". The condition is characterised by inflammation and fibrosis of the joint capsule, and early disease can be difficult to distinguish from other causes of a painful shoulder.[3]

Epidemiology

Frozen shoulder is estimated to affect 2–5% of the general population. It is most common between about 40 and 60 years of age and is more frequent in women (approximately 70% of presentations in older sources).[2][4] It has been called the "fifty-year shoulder" in China and Japan.

It is more common in people with diabetes. A meta-analysis of 13 cross-sectional studies estimated a prevalence of 13.4% (95% CI 10.2–17.2) in populations with diabetes, compared with roughly 2–5% in the general population.[5] Older sources report 10–22%; ranges depend on the study population and diagnostic criteria.

After one shoulder is affected, the contralateral shoulder is involved in a proportion of people (reported between 5% and 34%), and simultaneous bilateral involvement has been reported in about 14%.[4]

Clinical Presentation and Natural History

Presentation is typically an insidious onset of increasing pain with gradual loss of active and passive range of motion.[4] The most consistent finding is loss of external rotation with the arm by the side. Patients often report difficulty with grooming, dressing, overhead reaching and fastening items behind the back. Night pain and disturbed sleep are common.

Sleep. Sleep is often interrupted in the early and middle parts of the condition. Pain, poor sleep and low mood influence each other, so clinicians should ask about sleep and consider monitoring it with a tool such as the Pittsburgh Sleep Quality Index or the Medical Outcomes Study Sleep Scale (MOS-Sleep; 12 items covering sleep disturbance, adequacy, somnolence, quantity, snoring and awakening short of breath or with headache).

Phases. The condition is commonly described as progressing through three overlapping phases:[3][4][6]

Phase Typical features Commonly quoted duration
Freezing (painful) Gradual onset of pain at rest, sharp pain at end range, night pain About 2–9 months
Frozen (stiffening) Pain eases, progressive loss of GH motion, pain mainly at end range About 4–12 months
Thawing (resolution) Gradual improvement in functional range About 5–24 months

These phases and durations are descriptive conventions rather than firm natural history. A systematic review of studies with untreated comparison groups found only low-quality, contradictory evidence and concluded that the idea of a predictable progression to complete resolution without treatment is not supported.[7] Many patients recover well, but a substantial minority report persistent pain or restriction beyond three years. Duration varies widely between individuals, and diabetes is associated with a slower recovery.[4][7]

Aetiology and Risk Factors

The cause of primary frozen shoulder is not known, and the condition is probably multifactorial, with genetic, metabolic and environmental contributions.[1]

Primary and secondary frozen shoulder. Primary (idiopathic) frozen shoulder has no identifiable cause. Secondary frozen shoulder follows a known cause, predisposing factor or surgical event, such as surgery, stroke or injury. After injury, an altered movement pattern to protect painful structures may change shoulder motor control, reduce range and gradually stiffen the joint. Secondary frozen shoulder is usually divided into three subcategories: systemic causes (diabetes mellitus and other metabolic conditions), extrinsic causes (cardiopulmonary disease, cervical disc pathology, stroke, humeral fracture, Parkinson's disease) and intrinsic causes (rotator cuff pathology, biceps tendinopathy, calcific tendinopathy, acromioclavicular joint arthritis).[4]

Metabolic and systemic risk factors. Diabetes is the best established risk factor. A meta-analysis of six case-control studies estimated that people with diabetes had 3.69 times the odds (95% CI 2.99–4.56) of developing frozen shoulder, although unmeasured confounding limits this estimate.[5] Mendelian randomisation analysis within a UK Biobank genome-wide study supported type 1 diabetes as a likely causal risk factor and found no evidence that obesity is causal, which suggests that glycaemic rather than mechanical factors link diabetes and frozen shoulder.[8] Thyroid disorders (Hypothyroidism), Dupuytren's disease, Parkinson's disease and stroke are also associated with frozen shoulder.[1] Metabolic syndrome is a cluster of conditions that raises the risk of type 2 diabetes. People with diabetes tend to respond less well to treatment than people without it.[4] Clinicians should consider screening for diabetes and thyroid disease in patients with frozen shoulder.[2]

Genetic factors. Family and twin studies suggested a genetic contribution.[9] A genome-wide association study of 10,104 cases in the UK Biobank identified five genome-wide significant loci, with considerable overlap with loci associated with Dupuytren's disease.[8] These are associations that help explain biology; they have no current role in clinical diagnosis or management.

Other risk factors. Shoulder injury (a fall on an outstretched hand, direct impact, dislocation), shoulder or other surgery, and complex regional pain syndrome are commonly reported precipitants.[4]

Differential Diagnosis

Several conditions can present with similar pain and stiffness and may also coexist with frozen shoulder. These include osteoarthritis, acute calcific bursitis or tendinitis, rotator cuff pathology, Parsonage-Turner syndrome, a locked posterior dislocation and proximal humeral fracture.[4][10]

Condition Key distinguishing features
Shoulder osteoarthritis Active range of motion may be limited in abduction and external rotation, but passive range is typically less limited; flexion is often most affected (whereas it is usually least affected in frozen shoulder). Radiographs help rule out osseous pathology.
Acromioclavicular joint dysfunction High arc of pain, pain with cross-body adduction (scarf test), tenderness on palpation of the joint.
Bursitis Can resemble early frozen shoulder, with non-traumatic onset of severe pain. Passive range is usually larger than in frozen shoulder.
Parsonage-Turner syndrome Brachial plexus inflammation; no trauma, painful restriction, pain usually settles faster; weakness or atrophy appears several weeks after onset.
Rotator cuff pathology Restriction does not typically follow the capsular pattern; strength tests may be abnormal or normal early on. MRI or ultrasound can identify soft tissue lesions.
Posterior dislocation Associated with a specific traumatic event; inability to fully supinate the forearm with shoulder flexion raises suspicion.
Active muscle guarding (motor control dysfunction) A 2015 conference abstract reported that all participants with suspected frozen shoulder gained range under anaesthesia, suggesting that some presentations are not explained by capsular contracture alone.[11] A full paper from the same group examined this question further.[12]

Pathophysiology

Frozen shoulder is regarded as a fibroinflammatory disorder of the shoulder capsule. The disease process affects the antero-superior capsule, the axillary recess and the coracohumeral ligament, and the rotator interval is typically contracted. There has been continuing disagreement about whether the underlying process is primarily inflammatory, fibrotic or neurogenic. Current evidence suggests synovial inflammation followed by capsular fibrosis, with fibroblast proliferation, differentiation into myofibroblasts, and deposition of disorganised type III (and type I) collagen with imbalanced extracellular matrix turnover.[1][3][13]

Inflammatory cytokines, including interleukin-1, interleukin-6, HMGB1 and transforming growth factor-β (TGF-β), have been implicated in sustained inflammation and fibrosis, and matrix metalloproteinases are involved in remodelling.[1][13] This balance of aggressive fibrosis and impaired remodelling is thought to produce the stiff capsule and thickened ligaments.

Emerging evidence. The following findings are preliminary and mostly come from small tissue studies or laboratory work; they are not yet applied clinically.

A small study (four control and five frozen shoulder capsules) found higher IL-17A expression in frozen shoulder tissue, with T cells as a source and fibroblasts that responded more strongly to IL-17A.[14] A tissue expression study reported higher expression of several matrix metalloproteinases, IL6, TNF and collagen-related genes than in comparison shoulder conditions, suggesting that catabolic and anabolic processes occur at the same time.[15] Recent reviews describe early inflammatory signalling through NF-κB and JAK-STAT pathways, with TGF-β/Smad, Wnt/β-catenin and YAP/TAZ networks maintaining myofibroblast activity, which may explain persistent contracture after inflammation settles.[16] Genome-wide association findings (see Aetiology) and the overlap with Dupuytren's disease also point to shared fibrotic biology.[8]

Two bioinformatics studies applied machine-learning algorithms to publicly available gene-expression datasets and proposed candidate genes shared between frozen shoulder and osteoporosis, and between frozen shoulder and Dupuytren's disease.[17][18] These are exploratory computational findings that require laboratory and clinical validation. No validated artificial-intelligence tool for diagnosing or predicting the course of frozen shoulder was identified in this review.

Anatomical changes. Arthroscopic and pathological observations describe a small joint with loss of the axillary fold, a tight anterior capsule and mild to moderate synovitis.[3] Joint volume has been reported to fall from about 15–35 cm³ to 5–6 cm³, and capsular changes have been compared with those of Dupuytren's contracture. Thickening and fibrosis of the rotator interval contracts the GH ligaments, with the inferior glenohumeral ligament thought to be the most important.

Motor control. The restriction of movement may not be due to capsular contracture alone: muscle guarding and altered motor control are thought to contribute.[11][12]

Anatomical Considerations

Glenohumeral joint

The inferior glenohumeral ligament acts as a "hammock" at the bottom of the joint, with an anterior band, a posterior band and a less taut middle section (the pouch). If it tightens, accessory movement at the GH joint reduces. The capsule allows an estimated 2–3 mm of distraction and contributes little to stability on its own, but the rotator cuff tendons insert into it, so rotator cuff activity alters capsular tension. The capsule is also a proprioceptive organ: if it becomes tight, localised mechanoreceptors are stimulated, which may increase rotator cuff activity and, in turn, tension around the joint. Local neurovascular changes related to inflammation may also occur. See also biomechanics of the shoulder.

Assessment

Diagnosis is based on history and physical examination. No specific clinical test or gold standard exists, and imaging is generally used to exclude other pathology rather than confirm frozen shoulder.[2][6] A Delphi consensus identified clinical identifiers for early primary frozen shoulder grouped into pain and movement domains, plus an age component (over 35 years).[6] Global loss of active and passive range with end-range pain in all directions is the central finding.

Subjective assessment

Take a full past medical history (to screen for red flags, diabetes and thyroid disease) and history of the presenting problem. Typical features are night pain, pain with sudden or unguarded movements, difficulty lying on the affected side, and pain that may refer from the base of the skull to the hand. Reaching overhead or to the side (for example hanging clothes or fastening a seat belt), personal hygiene, dressing and hair brushing are commonly limited. Concomitant neck pain is common because cervical muscles overwork to compensate for lost shoulder motion.[4]

Observation and screening

Observe posture and scapular position; scapular winging may be visible. Perform an upper quarter screen and neurological screen (dermatomes, myotomes, reflexes) to exclude cervical or neurological pathology.[4]

Range of movement

Assess cervical, thoracic and shoulder range of motion with overpressure, plus rib mobility. Reduced flexion, abduction, external rotation and internal rotation are key signs, and scapular substitution commonly accompanies active motion.[4] In frozen shoulder, external rotation is typically more limited than abduction, which is more limited than internal rotation (the capsular pattern).[19]

Controversy about the capsular pattern. Cyriax proposed that the greatest limitation in passive external rotation, followed by abduction and then internal rotation, indicates a capsular pattern and arthritis of the joint, with other patterns suggesting a non-capsular lesion. The existence and diagnostic value of a consistent capsular pattern has been debated, but it remains a useful consideration in examination.[19]

Glenohumeral ligaments and accessory movements

Assessing accessory movement gives an indication of global joint stiffness, and should be compared with the contralateral side, remembering that the other shoulder may not be normal. Both quality and quantity of movement matter. At 60° abduction there is roughly equal tension across all GH ligaments, which gives an overall indication of stiffness. In frozen shoulder the anterior and inferior capsule tend to be the most restricted, although joint mobility is reduced in all directions.[19] Flexion, internal rotation and horizontal flexion are coupled with anterior-superior translation of the humeral head, and extension, external rotation and abduction with external rotation are coupled with posterior translation (see arthrokinematics).

Structure Function Assessment (patient supine, scapula stabilised, arm by side unless stated)
Superior GH ligament and coracohumeral ligament Superior: stabilises in adduction, limits external rotation and inferior translation. Coracohumeral: anterior portion limits extension, posterior portion limits flexion; also limits inferior and posterior translation Take passive external rotation to end range, apply an anterior glide to the humeral head (posterior and lateral bands). Adding 10° extension assesses the anterior and medial band of the coracohumeral ligament
Middle GH ligament Stabilises in adduction plus external rotation, and in about 45° abduction with external rotation 10° extension, external rotation to end range, then 45° abduction with an antero-medial glide in the plane of the scapula; compare sides
Inferior GH ligament, anterior band Stabilises in abduction and external rotation Move from the previous position towards 90° abduction with external rotation, apply an anterior-medial glide. Use caution if instability is suspected, as this resembles the apprehension test
Inferior GH ligament, posterior band Posterior stability in abduction and internal rotation 90° abduction, 10° extension, full internal rotation, gentle anterior-lateral glide
Posterior capsule Posterior restraint 90° flexion, full internal rotation, end-range horizontal adduction; apply an axial stress in a posterior-lateral direction in the plane of the scapula (similar to testing for posterior instability)

Tightening of the superior GH ligament and coracohumeral ligament can alter arthrokinematics, increasing anterior-superior translation in flexion and reducing the subacromial space.

Resisted tests

Resisted external rotation, internal rotation and abduction can be tested seated. Weakness relative to the unaffected side and significant muscle guarding may be present.[4] Consider the stage and irritability before resisted testing.

Functional and special tests

The shoulder shrug sign (inability to elevate the arm to 90° abduction without elevating the shoulder girdle) is associated with frozen shoulder, glenohumeral arthritis and massive cuff tears.[20] Three function-related tests (hand to neck, hand to scapula and hand to opposite scapula, the scarf test) have excellent reliability and moderate inter-test correlation, although a direct relationship with activities of daily living should not be assumed.[21] These tests need adequate elbow, scapulothoracic and thoracic mobility.

Function-related tests (Yang and Lin 2006): hand to neck (A), hand to scapula (B), hand to opposite scapula (C).[21]

Outcome Measures

Outcome measure Domain
Shoulder Pain and Disability Index (SPADI) Shoulder-specific pain and disability
Disabilities of the Arm, Shoulder and Hand (DASH) Upper limb function
American Shoulder and Elbow Surgeons (ASES) score Shoulder pain and function
Simple Shoulder Test (SST) Shoulder function
Penn Shoulder Score (PSS) Pain, satisfaction and function
NPRS, VAS Pain intensity
SF-36 Health-related quality of life
Pittsburgh Sleep Quality Index, MOS-Sleep Sleep

The psychometric properties of SPADI, DASH, ASES and SST are favourable across shoulder disorders, although a systematic review did not examine them specifically in frozen shoulder.[22]

Management

Principles

Recommendations on management are limited by low-quality evidence. An umbrella review concluded that non-surgical and rehabilitative interventions are effective for frozen shoulder, but found no evidence that one approach is more effective than another, and called for higher-quality trials with standardised intervention content.[23] A Cochrane review of manual therapy and exercise found limited and heterogeneous evidence; no trial compared manual therapy plus exercise with placebo or no treatment.[24] A systematic review of physiotherapy interventions (33 articles) graded some techniques as strongly recommended and others as moderately or mildly recommended, although the grading method was limited.[25]

In practice, treatment is individualised to the stage and irritability of the condition, the person's circumstances and the presence of conditions such as diabetes. Pain and sleep should be managed early.

Education and self-management

Education helps reduce frustration and supports adherence. Explain that stiffness usually improves over time, that full range may not return, and that recovery can be slow, particularly with diabetes. A home exercise programme that is easy to follow and carried out regularly is the foundation of rehabilitation. A network meta-analysis found that adding a simple home exercise programme to intra-articular corticosteroid gave additional mid-term pain benefit.[26] Relaxation, desensitisation, breathing exercises and graded distraction during aggravating movements may help in a condition that is painful, chronic and stressful.

Stage-based physiotherapy

Stage-based treatment is a clinical convention rather than a well-tested protocol, and relies mainly on older studies and expert consensus.[4][27]

Freezing (high irritability). The aim is pain relief and sleep support. Exercise is gentle, pain-free and low intensity, for example pendulum exercises, supported passive flexion, passive external rotation in about 40° abduction in the plane of the scapula, and active-assisted movement, held briefly (one to five seconds) in a pain-free range several times a day.[4] Heat before or during treatment may help with stretching.[4] Better outcomes were reported with simple pain-free exercise than with intensive therapy in this stage.[3] Grade I–II mobilisations are commonly used.

Frozen (stiffening). Mobilisation and stretching are progressed as irritability allows. Care is needed not to over-treat. Techniques include end-range mobilisation and mobilisation with movement. In one trial, mobilisation with movement was reported as more effective than stretching alone,[28] and end-range mobilisation was reported to be more effective than mid-range mobilisation for motion and function.[29] High-grade (Maitland grade III–IV) mobilisation produced greater improvements in passive abduction (3 and 12 months) and active and passive external rotation (12 months) than low-grade mobilisation in a randomised trial.[30] Posterior glide was reported to improve external rotation more than anterior glide in a small trial.[31] Positional stretching of the coracohumeral ligament has been reported only in a single case report. Dynamic splinting combined with physiotherapy was reported to improve active external rotation compared with either alone in a controlled cohort study.[32] These trials are small and often more than a decade old.

Thawing (resolution). As irritability falls, stretches can be held longer and performed more often, with higher-intensity stretching and strengthening (including scapular stabilisation and rotator cuff strengthening) to maintain gains and restore motor control.[4] Total end range time varies between people. Discharge decisions usually rest on pain reduction, a plateau in motion gains, and improved function and satisfaction rather than on range alone.[4]

Motor control. Muscle guarding and altered recruitment may contribute to restricted movement.[11][12] Goals are to restore normal recruitment patterns, balanced agonist and antagonist activity, appropriate effort for low loads, and control during isometric, concentric and eccentric activity.

Kinetic chain. Mobility and stability through the trunk and lower limbs support efficient shoulder movement during rehabilitation.


Active assisted end-range movements in flexion, external rotation and extension.
Active assisted end-range movements in internal rotation, horizontal adduction and flexion.
Coracohumeral ligament stretch.

Active assisted movements in elevation and external rotation with a cane.
Posterior mobilisations of the humeral head in supine.
Anterior mobilisations of the humeral head in supine.

Evidence for physiotherapy interventions

Intervention Evidence and effect Outcomes Certainty and limitations Key evidence
Education and home exercise Adding a simple home exercise programme to intra-articular corticosteroid gave additional mid-term benefit Pain, function Moderate; mostly studied as an adjunct to injection Challoumas 2020[26]
Structured physiotherapy plus steroid injection In secondary care, not inferior to manipulation under anaesthesia or arthroscopic capsular release; fewer serious adverse events than capsular release Function, quality of life, cost Moderate to high (pragmatic RCT, n = 503); applies to a specific 12-session protocol UK FROST, Rangan 2020[33]
Exercise (stretching, mobility) Pooled analysis of pre-post change found large improvements in ROM, function and pain, with larger ROM gains when joint mobilisation was added ROM, function, pain Low: pooled within-group change rather than controlled comparisons, high heterogeneity, unregistered protocol Lee 2023[34]
Manual therapy plus exercise Cochrane review found no trial against placebo or no treatment, and heterogeneous comparisons; may be less effective than steroid injection in the short term Pain, function, ROM Low; this review is from 2014 Page 2014[24]
Joint mobilisation High-grade mobilisation improved range more than low-grade at 3 to 12 months in one RCT; further trials on type and direction are small ROM, disability Low; small older trials Vermeulen 2006[30]
Scapulothoracic physiotherapy Pooled moderate to large effect favouring scapulothoracic interventions; no significant between-group difference in pain Shoulder function, pain Low to moderate; considerable heterogeneity 2026 systematic review and meta-analysis[35]
Therapeutic ultrasound No evidence of greater improvement in abduction than placebo; findings for other outcomes inconsistent Pain, ROM, disability Low; five studies in meta-analysis Sung 2022[36]
Extracorporeal shockwave therapy Authors report evidence of pain reduction in chronic frozen shoulder with type 2 diabetes Pain Low; seven studies, one subpopulation Phys Ther 2025[37]
Intra-articular corticosteroid plus physiotherapy Corticosteroid gave the greatest short-term benefit (about 1 VAS point for pain versus no treatment or physiotherapy), lasting up to 6 months in early disease. Steroid plus physiotherapy was most beneficial in the freezing phase Pain, function Moderate; benefit greatest for symptoms under 1 year Challoumas 2020[26]; Zhang 2021[38]
Hydrodilatation with or without physiotherapy Conflicting: a 2023 meta-analysis found a transient disability benefit over steroid alone (SMD 0.24) and no difference from manipulation, capsular release or general physiotherapy. A 2026 review of 44 RCTs found greater external rotation at 12 weeks (low certainty) but no difference in pain or SPADI and concluded it does not support meaningful benefit over steroid alone or as an adjunct to physiotherapy Pain, disability, ER Low; variable technique, volume and capsule rupture Poku 2023[39]; 2026 review[40]

Modalities and other approaches. Heat and cold have been used with exercise; hot and cold packs before and after exercise were reported to improve outcomes in a randomised study that also used steroid injection.[41] A retrospective cohort study reported that ultrasound, massage, iontophoresis and phonophoresis were associated with a reduced likelihood of improvement,[42] and a 2022 meta-analysis of ultrasound found no consistent benefit over placebo.[36] Acupuncture, dry needling, kinesiology taping and proprioceptive neuromuscular facilitation are used clinically, but their effects were not appraised in this update (a 2019 meta-analysis concluded that PNF reduced pain and improved range and function).[43] A Cochrane review of physiotherapy for shoulder pain found no evidence of an effect of ultrasound in frozen shoulder.[44]

Injection and interventional management

Corticosteroid injection. Corticosteroid injection aims to reduce synovitis in the early painful stage. A network meta-analysis of 65 studies found intra-articular corticosteroid superior to other treatments in the short term for pain (mean difference about −1.0 VAS point versus no treatment, −1.1 versus physiotherapy) and function, with benefit lasting up to 6 months in people with symptoms under one year.[26] Another network meta-analysis found that steroid plus physiotherapy was most beneficial in the freezing phase.[38] Earlier reviews found a short-term advantage at 4–6 weeks, with little difference from physiotherapy by 12 weeks.[45][46] Subacromial and intra-articular injection have been reported to have similar effects. Contraindications include infection, coagulopathy and uncontrolled diabetes, and corticosteroid can transiently raise blood glucose in people with diabetes. Injection should be offered together with exercise.[26]

Hydrodilatation (distension arthrography). Hydrodilatation instils a large volume (often about 30 mL) of saline, steroid and local anaesthetic into the joint under imaging guidance. The role of capsular rupture is unclear, and systematic reviews disagree on benefit over corticosteroid alone (see table).[39][40][47]

Other injections and oral drugs. NSAIDs have been used, but high-level evidence for effectiveness is lacking.[3][46] Oral corticosteroids have short-term benefits only and are not recommended routinely.[48] Suprascapular nerve block has been used for pain, with some evidence of benefit but unclear mechanism.[46]

Surgical management

In UK FROST, neither manipulation under anaesthesia (MUA) nor arthroscopic capsular release was superior to early structured physiotherapy plus steroid injection in secondary care. Capsular release had more serious adverse events, and MUA had the highest probability of being cost-effective.[33] Surgical options are generally considered after at least about 6 months of unsuccessful conservative treatment, although there are no studies establishing the best time to switch.[4][46]

Manipulation under anaesthesia. MUA involves controlled end-range positioning of the humerus under a brachial plexus block, now using short rather than long lever arms. Risks include fracture, dislocation, nerve injury, rotator cuff tear and haemarthrosis.[4][49] Contraindications include previous fracture or dislocation, significant bone loss and inability to follow post-procedure care. Translation mobilisation under anaesthesia has been proposed as a safer alternative, but evidence is limited.[4]

Arthroscopic capsular release. This selectively releases the contracted capsule (most commonly the rotator interval and coracohumeral ligament) and avoids some MUA complications. It is the most widely used procedure for refractory frozen shoulder and is often favoured in diabetes and in post-operative or post-fracture frozen shoulder, although direct comparisons are few.[4][46]

Post-surgical considerations. Physiotherapy after surgery should be started promptly and individualised.[33] Consider nerve integrity near portal sites (axillary, suprascapular, musculocutaneous, radial, median, ulnar and long thoracic nerves and the brachial plexus), because sudden restoration of movement after months of stiffness may irritate nerves, and assess the cervical spine and nerve root mobility.

Clinical Considerations and Red Flags

Features that should prompt consideration of systemic or other pathology include fever or chills, unexplained severe pain, a history of cancer or suspicion of metastatic disease, multiple joint involvement, shortness of breath or severe cough, and any suspicion of systemic illness such as tuberculosis or inflammatory arthritis (see flag system). Frozen shoulder after stroke and in association with complex regional pain syndrome needs specific consideration.

Clinical Bottom Line

No intervention has clearly superior evidence, and treatment should match the stage, irritability, comorbidity and preferences of the individual. Early pain control and sleep support, education and a home programme are core. For people with high pain levels in early disease, corticosteroid injection combined with structured physiotherapy has the strongest comparative evidence.[33][26][38] Surgical options are generally reserved for people who remain disabled despite conservative care, with MUA and capsular release producing outcomes comparable to physiotherapy plus injection in the largest trial.[33] The cause and natural history remain incompletely understood, and the aim in clinical practice is to focus on the quality of movement and function rather than range alone.

References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 Millar NL, Meakins A, Struyf F, et al. Frozen shoulder. Nat Rev Dis Primers 2022;8:59.
  2. ↑ 2.0 2.1 2.2 2.3 Ramirez J. Adhesive capsulitis: diagnosis and management. Am Fam Physician 2019;99(5):297-300.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Dias R, Cutts S, Massoud S. Frozen shoulder. BMJ 2005;331:1453-6.
  4. ↑ 4.00 4.01 4.02 4.03 4.04 4.05 4.06 4.07 4.08 4.09 4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17 4.18 4.19 4.20 4.21 Kelley MJ, McClure PW, Leggin BG. Frozen shoulder: evidence and a proposed model guiding rehabilitation. J Orthop Sports Phys Ther 2009;39:135-48.
  5. ↑ 5.0 5.1 Dyer BP, et al. Diabetes as a risk factor for the onset of frozen shoulder: a systematic review and meta-analysis. BMJ Open 2023;13:e062377.
  6. ↑ 6.0 6.1 6.2 Walmsley S, Rivett DA, Osmotherly PG. Adhesive capsulitis: establishing consensus on clinical identifiers for stage 1 using the Delphi technique. Phys Ther 2009;89:906-17.
  7. ↑ 7.0 7.1 Wong CK, et al. Natural history of frozen shoulder: fact or fiction? A systematic review. Physiotherapy 2017;103:40-47. doi:10.1016/j.physio.2016.05.009.
  8. ↑ 8.0 8.1 8.2 Green HD, et al. A genome-wide association study identifies 5 loci associated with frozen shoulder and implicates diabetes as a causal risk factor. PLoS Genet 2021.
  9. ↑ Hakim AJ, Cherkas LF, Spector TD, MacGregor AJ. Genetic associations between frozen shoulder and tennis elbow: a female twin study. Rheumatology (Oxford) 2003;42:739-42.
  10. ↑ Kline CM. Adhesive capsulitis: clues and complexities. JAMA 2007.
  11. ↑ 11.0 11.1 11.2 Hollmann L, Halaki M, Haber M, Herbert R, Dalton S, Ginn K. Determining the contribution of active stiffness to reduced range of motion in frozen shoulder. Physiotherapy 2015;101:e585.
  12. ↑ 12.0 12.1 12.2 Hollmann L, Halaki M, Kamper SJ, Haber M, Ginn KA. Does muscle guarding play a role in range of motion loss in patients with frozen shoulder? Musculoskelet Sci Pract 2018;37:64-68.
  13. ↑ 13.0 13.1 Kraal T, Lubbers J, van den Bekerom MPJ, et al. The puzzling pathophysiology of frozen shoulders: a scoping review. J Exp Orthop 2020;7. doi:10.1186/s40634-020-00307-w.
  14. ↑ Akbar M, et al. Translational targeting of inflammation and fibrosis in frozen shoulder: molecular dissection of the T cell/IL-17A axis. Proc Natl Acad Sci U S A 2021;118(39):e2102715118.
  15. ↑ Nishimoto H, et al. Characteristics of gene expression in frozen shoulder. BMC Musculoskelet Disord 2022;23:811. doi:10.1186/s12891-022-05762-3.
  16. ↑ Signaling pathway networks in frozen shoulder fibrosis: from inflammatory initiation to fibrotic progression. Front Med 2026. doi:10.3389/fmed.2026.1883535.
  17. ↑ Identification of common genetic features and pathways for osteoporosis with frozen shoulder by integrated bioinformatics analysis and machine learning. PMID 41276899.
  18. ↑ Exploring the potential relationship between frozen shoulder and Dupuytren's disease through bioinformatics analysis and machine learning. PMC10500125.
  19. ↑ 19.0 19.1 19.2 Mitsch J, Casey J, McKinnis R, Kegerreis S, Stikeleather J. Investigation of a consistent pattern of motion restriction in patients with adhesive capsulitis. J Man Manip Ther 2004;12:153-9.
  20. ↑ Jia X, Ji JH, Petersen SA, Keefer J, McFarland EG. Clinical evaluation of the shoulder shrug sign. Clin Orthop Relat Res 2008;466(11):2813-9.
  21. ↑ 21.0 21.1 Yang J, Lin J. Reliability of function-related tests in patients with shoulder pathologies. J Orthop Sports Phys Ther 2006;36:572-6.
  22. ↑ Roy J, MacDermid J, Woodhouse L. Measuring shoulder function: a systematic review of four questionnaires. Arthritis Rheum 2009;61(5):623-32.
  23. ↑ de Sire A, Agostini F, Bernetti A, et al. Non-surgical and rehabilitative interventions in patients with frozen shoulder: umbrella review of systematic reviews. J Pain Res 2022;15:2449-64.
  24. ↑ 24.0 24.1 Page MJ, Green S, Kramer S, et al. Manual therapy and exercise for adhesive capsulitis (frozen shoulder). Cochrane Database Syst Rev 2014;(8):CD011275.
  25. ↑ Nakandala P, Nanayakkara I, Wadugodapitiya S, Gawarammana I. The efficacy of physiotherapy interventions in the treatment of adhesive capsulitis: a systematic review. J Back Musculoskelet Rehabil 2021;34:195-205.
  26. ↑ 26.0 26.1 26.2 26.3 26.4 26.5 Challoumas D, Biddle M, McLean M, Millar NL. Comparison of treatments for frozen shoulder: a systematic review and meta-analysis. JAMA Netw Open 2020;3(12):e2029581.
  27. ↑ Kelley MJ, et al. Shoulder pain and mobility deficits: adhesive capsulitis clinical practice guidelines. J Orthop Sports Phys Ther 2013;43(5):A1-A31.
  28. ↑ Doner G, Guven Z, Atalay A, Celiker R. Evaluation of Mulligan's technique for adhesive capsulitis of the shoulder. J Rehabil Med 2013;45(1):87-91.
  29. ↑ Yang JL, Chang CW, Chen SY, Wang SF, Lin JJ. Mobilization techniques in subjects with frozen shoulder syndrome: randomized multiple-treatment trial. Phys Ther 2007;87:1307-15.
  30. ↑ 30.0 30.1 Vermeulen HM, Rozing PM, Obermann WR, le Cessie S, Vliet Vlieland TP. Comparison of high-grade and low-grade mobilization techniques in the management of adhesive capsulitis of the shoulder: randomized controlled trial. Phys Ther 2006;86:355-68.
  31. ↑ Johnson AJ, Godges JJ, Zimmerman GJ, Ounanian LL. The effect of anterior versus posterior glide joint mobilization on external rotation range of motion in patients with shoulder adhesive capsulitis. J Orthop Sports Phys Ther 2007;37:88-99.
  32. ↑ Gaspar P, Willis B. Adhesive capsulitis and dynamic splinting: a controlled cohort study. BMC Musculoskelet Disord 2009;10:111.
  33. ↑ 33.0 33.1 33.2 33.3 33.4 Rangan A, et al. Management of adults with primary frozen shoulder in secondary care (UK FROST): a multicentre, pragmatic, three-arm, superiority randomised clinical trial. Lancet 2020;396(10256):977-89.
  34. ↑ Lee JH, Jeon HG, Yoon YJ. Effects of exercise intervention (with and without joint mobilization) in patients with adhesive capsulitis: a systematic review and meta-analysis. Healthcare 2023;11:1504. doi:10.3390/healthcare11101504.
  35. ↑ Effects of scapulothoracic physical therapy on shoulder function in adhesive capsulitis: a systematic review and meta-analysis. PMID 42631081 (PROSPERO CRD420251126688).
  36. ↑ 36.0 36.1 Sung JH, Lee JM, Kim JH. The effectiveness of ultrasound deep heat therapy for adhesive capsulitis: a systematic review and meta-analysis. Int J Environ Res Public Health 2022;19:1859. doi:10.3390/ijerph19031859.
  37. ↑ Extracorporeal shock wave therapy for chronic adhesive capsulitis in type 2 diabetics: a systematic review with meta-analysis. Phys Ther 2025;105(7):pzaf074.
  38. ↑ 38.0 38.1 38.2 Zhang J, Zhong S, Tan T, et al. Comparative efficacy and patient-specific moderating factors of nonsurgical treatment strategies for frozen shoulder: an updated systematic review and network meta-analysis. Am J Sports Med 2021.
  39. ↑ 39.0 39.1 Poku D, et al. Efficacy of hydrodilatation in frozen shoulder: a systematic review and meta-analysis. Br Med Bull 2023;147(1):121.
  40. ↑ 40.0 40.1 Hydrodilatation for adhesive capsulitis: a systematic review exploring efficacy and optimal technique. PMID 42527106.
  41. ↑ Bal A, Eskioglu E, Gulec B, Aydog E, Gurcay E, Cakci A. Effectiveness of corticosteroid injection in adhesive capsulitis. Clin Rehabil 2008;22:503-12.
  42. ↑ Jewell DV, Riddle DL, Thacker LR. Interventions associated with an increased or decreased likelihood of pain reduction and improved function in patients with adhesive capsulitis: a retrospective cohort study. Phys Ther 2009;89:419-29.
  43. ↑ Tedla JS, Sangadala DR. Proprioceptive neuromuscular facilitation techniques in adhesive capsulitis: a systematic review and meta-analysis. J Musculoskelet Neuronal Interact 2019;19(4):482-91.
  44. ↑ Green S, Buchbinder R, Hetrick SE. Physiotherapy interventions for shoulder pain. Cochrane Database Syst Rev 2003 (update 2010).
  45. ↑ Buchbinder R, Green S, Youd JM. Corticosteroid injections for shoulder pain. Cochrane Database Syst Rev 2009;1:CD004016.
  46. ↑ 46.0 46.1 46.2 46.3 46.4 Neviaser AS, Hannafin JA. Adhesive capsulitis: a review of current treatment. Am J Sports Med 2010;38:2346-56.
  47. ↑ Rymaruk S, Peach C. Indications for hydrodilatation for frozen shoulder. EFORT Open Rev 2017;2(11):462-8.
  48. ↑ Buchbinder R, Green S, Youd JM, Johnston RV. Oral steroids for adhesive capsulitis. Cochrane Database Syst Rev 2009;1:CD006189.
  49. ↑ Boyles RE, Flynn TW, Whitman JM. Manipulation following regional interscalene anesthetic block for shoulder adhesive capsulitis: a case series. Man Ther 2005;10:164-71.
Retrieved from "https://www.physio-pedia.com/w/index.php?title=Frozen_Shoulder&oldid=383648"