Frozen shoulder, traditionally known as adhesive capsulitis, is a chronic condition characterised by synovial inflammation and progressive capsular fibrosis. While the term "adhesive" was coined due to early observations of the capsule appearing "stuck" to the humerus, recent arthroscopic evidence has refuted the presence of true intra-articular adhesions, leading some experts to prefer the term "contracted shoulder" or "fibrotic capsulitis".
Core Pathophysiological Mechanisms¶
The pathogenesis involves a complex biological cascade where an initial inflammatory trigger---possibly systemic, extrinsic, or intrinsic---leads to a pro-fibrotic environment.
Knowledge Check
Frozen shoulder involves synovial inflammation and progressive capsular fibrosis, not true adhesions. The term "adhesive" is actually a misnomer as arthroscopic evidence shows no true intra-articular adhesions.
• Inflammation and Synovitis: In the early stages, the shoulder exhibits a hypervascular, hypertrophic synovitis. This is marked by an increase in pro-inflammatory cytokines (such as IL-1, IL-6, and TNF-alpha) and growth factors (like TGF-beta and PDGF).
• Fibrosis and Myofibroblasts: The inflammatory phase triggers a fibroproliferative response. Fibroblasts transform into myofibroblasts, pathognomonic contractile cells that lay down a dense, disorganized matrix of type I and type III collagen. "swimmer's shoulder," • Matrix Remodelling Imbalance: There is a significant disruption in the balance between matrix metalloproteinases (MMPs), which degrade the extracellular matrix, and their inhibitors (TIMPs). This imbalance prevents the breakdown of excess collagen, leading to permanent tissue thickening and contracture.
Anatomical and Structural Changes
The disease process does not affect the entire joint equally but specifically targets the anterosuperior joint capsule and related structures.
• Rotator Interval and CHL: The rotator interval (the triangular space between the supraspinatus and subscapularis) and the coracohumeral ligament (CHL) are the primary sites of contracture. Thickening of the CHL is a hallmark finding and is largely responsible for the characteristic loss of passive external rotation.
Knowledge Check
Thickening of the coracohumeral ligament is a hallmark finding in frozen shoulder and is largely responsible for the characteristic loss of passive external rotation.
• Joint Volume Reduction: Contracture and thickening of the capsule, particularly the axillary fold, lead to a significant reduction in joint volume---often from a normal 15--35 mL down to less than 5--10 mL.
Neurovascular and Metabolic Factors¶
Current knowledge highlights the role of both new blood vessels and nerves in the symptoms of frozen shoulder.
• Neoangiogenesis and Neoinnervation: Clinical pain, particularly at night, is attributed to hypervascularity (neoangiogenesis) and an increase in nerve fibre density (neoinnervation) within the capsule. Elevated levels of neuronal proteins like Substance P and CGRP contribute to chronic pain and hyperalgesia.
• Diabetes and Glycation: The strong link to diabetes mellitus is thought to be driven by the accumulation of advanced glycation end products (AGEs). High glucose levels lead to collagen cross-linking, which reduces tissue compliance and makes the capsule more resistant to degradation.
Emerging Pathophysiological Concepts¶
Recent research has introduced additional dimensions to our understanding of the condition:
• Muscle Guarding: Small-scale studies have demonstrated that some patients experience a substantial increase in range of motion under general anaesthesia, suggesting that active muscle guarding or "fear-based brain-induced movement limitation" may be as significant as structural contracture in some individuals.
• Systemic Low-Grade Inflammation: There is an increasing hypothesis that frozen shoulder may be a symptom of metabolic syndrome or systemic low-grade inflammation, rather than a purely local shoulder issue.
• Association with Dupuytren\'s: The histology of the frozen shoulder capsule is nearly indistinguishable from Dupuytren's disease, suggesting they may share common biochemical pathways of fibromatosis
Historical tour of Frozen shoulder¶
The historic understanding of frozen shoulder has evolved over 150 years, moving from vague descriptions of general stiffness to a sophisticated understanding of specific cellular and molecular changes within the joint capsule. There's every chance that we still don't really know what's going on with frozen shoulder, so it's worthwhile remembering our historic understanding of this problem so we can put our current understanding into perspective.
Late 19th and Early 20th Centuries: Vague Origins¶
• 1872: The French surgeon Duplay provided the first description of the condition, naming it "peri-arthritis scapulo-humerale". He attributed the disabling pain and restriction to inflammation of the subacromial bursa rather than the joint itself.
• 1907: Following the advent of radiographs, some clinicians, such as Baer, briefly attributed the symptoms to newly visible calcific deposits in the shoulder.
The 1930s: Defining the "Enigma"¶
• 1934: Ernest Codman coined the term "frozen shoulder" to replace "adherent subacromial bursitis". He established the classic diagnostic criteria: insidious onset, pain near the deltoid insertion, inability to sleep on the affected side, and a profound loss of elevation and external rotation despite normal X-rays. Codman famously described the condition as "difficult to define, difficult to treat and difficult to explain".
The 1940s: The Adhesive Theory¶
• 1943: Lippmann argued that the condition was actually "bicipital tenosynovitis," caused by the long head of the biceps tendon becoming stuck to its sheath.
• 1945: Neviaser introduced the term "adhesive capsulitis". Based on surgical observations, he described the joint capsule as appearing "stuck" to the humeral head like adhesive plaster on bare skin.
Mid-20th Century: Classification and Staging¶
• 1965: Andrén and Lundberg first described joint distension (hydrodilatation) as a treatment method, aiming to rupture the contracted capsule with high-volume fluid.
• 1969: Lundberg published a landmark study categorising the condition into "primary" (idiopathic) and "secondary" (associated with a known cause). He was the first to report that the pathology was fibrosis and fibroplasia rather than a significant inflammatory cell infiltrate, noting its similarity to Dupuytren's contracture.
• 1975: Reeves established the traditional three-phase natural history: the painful (freezing) phase, the stiff (frozen) phase, and the recovery (thawing) phase.
Late 20th Century: Arthroscopic and Cellular Insights¶
• 1979: Conti provided the first description of arthroscopic resection of the shoulder capsule.
• 1987: Neviaser and Neviaser proposed a refined four-stage classification system based on correlating physical examinations with arthroscopic findings.
• 1989: Ozaki and colleagues identified the coracohumeral ligament (CHL) and the rotator interval as the primary anatomical sites of contracture.
• 1995: Bunker and Anthony performed immunocytochemical analyses, proving that the cell population in the capsule consists of fibroblasts and myofibroblasts. They also reported that up to 58% of patients had concurrent Dupuytren's disease.
The 21st Century: Molecular and Neurological Frontiers¶
• 2011/2014: Major clinical bodies, including the American Shoulder and Elbow Surgeons (ASES) and ISAKOS, reached a consensus to prefer the term "frozen shoulder" for idiopathic cases and recommended abandoning the term "adhesive" because arthroscopic evidence has repeatedly shown a lack of true intra-articular adhesions.
• 2015--2018: Research by Hollmann and Ginn introduced the "muscle guarding" hypothesis, demonstrating that some patients regain significant range of motion under general anaesthesia, suggesting that the brain and central nervous system play a larger role in restriction than physical contracture alone.
• Present Day: Contemporary understanding focuses on a pro-inflammatory cytokine cascade (e.g., TGF-beta, IL-1, IL-6) and metabolic links to diabetes and systemic low-grade inflammation
Clinical course of frozen shoulder¶
The clinical course of frozen shoulder (adhesive capsulitis) classically follows three overlapping phases: the freezing phase (inflammation and worsening pain), the frozen phase (maximal stiffness with receding pain), and the thawing phase (gradual recovery of motion). While traditionally described as a self-limiting condition lasting 12 to 42 months, recent evidence suggests that up to 40--50% of patients may experience residual pain or functional deficits for many years.
Knowledge Check
Recent evidence shows that up to 40-50% of frozen shoulder patients may experience residual pain or functional deficits for many years, challenging the traditional view of it being simply self-limiting.
Nonsurgical Treatment Options¶
Supportive Management and Supervised Neglect:¶
◦ Pros/Outcomes: Focuses on patient education and painless exercise, which has been shown to produce better functional results than intensive physical therapy in some studies.
◦ Cons: Can involve a protracted duration of disability (often years) before resolution occurs.
Oral Medications (NSAIDs and Steroids):¶
◦ Pros/Outcomes: NSAIDs provide short-term pain relief but do not affect range of motion (ROM). A short course of oral steroids provides significant rapid relief for pain and function, but benefits typically dissipate after 6 weeks.
◦ Cons: Oral steroids carry risks of systemic side effects (e.g., blood sugar spikes) and can cause rebound pain upon withdrawal.
Intra-articular Corticosteroid Injections (CSI):¶
◦ Pros/Outcomes: Regarded as the most effective early intervention (especially for symptoms \<1 year), providing superior pain relief and functional improvement for up to 6 months.
◦ Cons: They do not alter the long-term natural history or final ROM outcomes.
Physical Therapy and Exercise:¶
◦ Pros/Outcomes: Physiotherapy is essential for maintaining motion, especially when combined with CSI. Group exercise or home programs may be as effective as individual therapy.
◦ Cons: Intensive stretching can be counterproductive in the early inflammatory phase, potentially exacerbating symptoms.
Invasive and Surgical Options¶
Hydrodilatation (Joint Distension):¶
◦ Pros/Outcomes: Uses fluid pressure to expand the capsule; reported to yield higher patient satisfaction and better functional scores at 6 months compared to manipulation under anaesthesia (MUA).
◦ Cons: The procedure can be painful and its long-term superiority over simple CSI remains unproven.
Manipulation Under Anaesthesia (MUA):¶
◦ Pros/Outcomes: Provides a reliable mechanical release for refractory cases, often shortening the recovery time to work.
◦ Cons: Carries risks of iatrogenic damage, including humeral fractures, rotator cuff tears, and labral lesions.
Arthroscopic Capsular Release (ACR):¶
◦ Pros/Outcomes: Allows a precise, controlled release of specific structures (like the coracohumeral ligament) and visual confirmation of the diagnosis. It can lead to immediate relief of night pain.
◦ Cons: It is the most invasive and expensive option, associated with higher risks of serious adverse events (e.g., joint instability or nerve injury) than MUA or PT.
Summary of Reported Outcomes¶
The UK FROST trial, the largest study of its kind, found that none of the major interventions (ACR, MUA, or structured physio with CSI) were clinically superior at 12 months; all led to substantial recovery, with MUA found to be the most cost-effective. Ultimately, treatment selection should be tailored to the patient\'s tissue irritability and stage of disease
Knowledge Check
The UK FROST trial found that ACR, MUA, and structured physio with CSI all led to substantial recovery with no clear clinical superiority at 12 months, though MUA was the most cost-effective option.
Treatment Option Pros Cons Reported **Suitable **Unsuitable Outcomes Patients** Patients**
Supervised Neglect Avoids complications Protracted duration 89--94% reach Patients Patients requiring
/ Benign Neglect of aggressive of disability; pain near-normal, preferring rapid restoration
therapy; very low may persist for painless non-invasive of motion for work
cost. years. function paths; or sport.
within 2 low-demand
years. individuals.
Oral Medications Short-term relief of No improvement in Decreases pain Early Patients needing (NSAIDs) pain at rest and range of motion symptoms from inflammatory improvements in during sleep. (ROM); risk of baseline but (Stage 1) joint mobility or nausea. does not alter patients for those with gastric the disease analgesia. sensitivity. course.
Oral Rapid reduction in Benefits typically Short-term Patients in Diabetics (blood
Corticosteroids pain and improvement dissipate after 6 symptomatic early stages sugar risk);
in function within weeks; risk of relief; requiring patients requiring
first 3 weeks. systemic side potential for fast, long-term
effects. "rebound short-term solutions.
pain" after pain relief.
withdrawal.
Intra-articular Regarded as the most Does not alter Superior Symptoms \<1 Late-stage patients
Steroid Injections effective long-term ROM or short-term year; early with stiffness but
(CSI) early-stage natural history; results (\<6 inflammatory minimal pain (Stage
intervention for potential for months); stage (Stage 1 ¾).
pain and function. transient facial accuracy or 2).
flushing. improved by
imaging
guidance.
Physical Therapy Essential for Intensive Supervised Frozen and High-irritability
(PT) / Exercise maintaining motion stretching can be exercise is thawing phase patients if therapy
and improving counterproductive often more patients; exceeds pain
function when and exacerbate pain effective than those capable limits.
combined with CSI. in early phases. unsupervised of regular
home programs. compliance.
Hydrodistension Less invasive than Procedure can be Higher patient Refractory Patients in early
(Hydrodilatation) surgery; increases painful; long-term satisfaction cases failing high-irritability
capsular volume superiority over and Constant standard PT; stages; may not be
directly. CSI is unproven. scores at 6 patients in available in all
months the "frozen" settings.
compared to phase.
MUA.
Manipulation Under Reliable mechanical Risks of humeral Satisfactory Refractory Elderly or Anaesthesia (MUA) release of fracture, rotator outcomes (pain Stage 2 cases osteoporotic refractory cuff tears, and relief and failing 6 patients (fracture stiffness. labral lesions. ROM) in months of risk); patients 75--97% of nonoperative with previous patients. care. fractures.
Arthroscopic Precise, controlled Most invasive and Significant Recalcitrant Patients in the
Capsular Release release; visual expensive; risk of and rapid cases (6--12 acutely painful
(ACR) confirmation of joint instability improvements months); pre-adhesive stage
diagnosis/staging. or nerve injury. in night pain diabetic (Stage 1).
and ROM within patients who
the first often fail
week. MUA.
Key Clinical Considerations¶
The UK FROST Trial: This major study found that no intervention was clinically superior at 12 months; however, MUA was the most cost-effective for secondary care.
• Diabetes Impact: Diabetic patients generally have poorer responses to treatment and may require more invasive interventions like ACR because they respond poorly to MUA alone.
• Stage-Based Approach: Success depends on matching the treatment to the tissue irritability level. High irritability favours pain reduction (CSI/NSAIDs), while low irritability permits more aggressive stretching and surgery
Corticosteroids in frozen shoulder¶
Corticosteroid use in patients with frozen shoulder (adhesive capsulitis) requires careful consideration due to potential systemic and local side effects, with diabetes mellitus representing a significant area for precaution.
General Precautions and Side Effects¶
While both oral and intra-articular (IA) corticosteroids provide effective short-term symptomatic relief, they carry specific risks:
• Systemic Risks: Oral corticosteroids, especially in high doses or prolonged courses, are associated with risks such as avascular necrosis (AVN) of the femoral head, endocrine disruption, and adverse interactions with other medications.
• Rebound Pain: Patients frequently experience a "rebound effect" or "flare-up"---a temporary worsening of shoulder pain---following the cessation of an oral steroid course or the withdrawal of the medication\'s effects.
• Local Injection Reactions: Common transient side effects of intra-articular injections include facial flushing (reported in 2.3% to 15% of patients), dizziness, nausea, and vasovagal episodes. Some patients may experience prolonged pain at the injection site.
• Infection Risk: There is a small but serious risk of septic joint arthritis or infection following an injection. Consequently, many surgeons advise against performing surgery (such as capsular release) within three months of the last steroid injection.
Special Reference to Diabetic Patients¶
Diabetes is the most notable co-morbidity for frozen shoulder, and steroid use in this population necessitates strict monitoring:
• Glycaemic Control: Corticosteroids can significantly impair blood glucose control. They often cause a transient increase in serum glucose levels, particularly within the first 24 hours after an intervention.
• Acute Hyperglycaemia: While some shoulder-specific data suggest mean glucose changes may be minimal in well-controlled patients, evidence from other joint injections indicates that acute hyperglycaemia can persist for two to three days.
• Refractory Response: Diabetic patients often have a poorer or more protracted response to non-operative treatments, including steroids, and are more likely to require surgical intervention for persistent deficits.
• Formulation Preference: In diabetic subgroups, triamcinolone acetonide has been reported to yield superior pain relief compared to methylprednisolone acetate.
Contraindications¶
Corticosteroid treatments are generally avoided or strictly limited in the following scenarios:
• Pregnancy and Breastfeeding: Pregnant and breastfeeding mothers are typically excluded from corticosteroid treatment protocols in clinical settings.
• Active Infection: Steroids are contraindicated in the presence of existing joint or systemic infections.
• Type 1 Diabetes: Severe frozen shoulder in patients with type 1 diabetes may be considered a contraindication for certain steroid-based treatments, particularly if glycaemic control is unstable.
• Medical Instability: Steroids should be avoided in patients with other unstable conditions, such as severe asthma or recent myocardial infarction, depending on the route of administration
Explaining the condition to your patient¶
Frozen shoulder, also known as adhesive capsulitis, is a common but debilitating condition that results in a lot of shoulder pain and a progressive loss of both active and passive movement. It typically affects people between the ages of 40 and 60, occurring more slightly more frequently in women and in individuals with medical conditions such as diabetes or thyroid disorders.
What's wrong?¶
In a healthy shoulder, the joint is surrounded by a flexible bag of connective tissue called the capsule. In a frozen shoulder, this capsule becomes inflamed and gradually thickens and scars, a process called fibrosis. This causes the capsule to tighten and contract, significantly reducing the space inside the joint and making movement difficult and painful.
While the term "adhesive" was historically used because the capsule was thought to be "stuck" to the bone like plaster on skin, modern evidence shows there are no actual adhesions (glue-like stickiness); rather, the entire capsule has simply become tight and thick. The hallmark sign of this condition is a severe loss of external rotation (the ability to turn your arm outwards), which is caused by the tightening of a specific ligament at the front of the shoulder.
What's going to happen?¶
Frozen shoulder is traditionally described as a self-limiting condition meaning that while it will get worse during its course, eventually things will reverse and improve. Usually it progresses through three overlapping stages:
• The Freezing Phase: Pain is the dominant feature, often becoming severe at night and making it impossible to sleep on the affected side. Movement starts to become limited during this phase.
• The Frozen Phase: The intense pain may begin to ease, but stiffness is now the primary problem. The shoulder feels locked, and reaching overhead or behind the back becomes very difficult.
• The Thawing Phase: The stiffness gradually begins to resolve, and your range of motion slowly improves.
The recovery process is usually slow, often taking between one and three years. While most patients eventually experience a significant improvement in function, up to 40--50% of people may be left with some degree of mild residual stiffness or pain many years later.
What can be done?¶
Treatment is usually tailored to the stage of the condition and your level of "tissue irritability" (how easily your pain is triggered).
• Early Care: In the early, painful stages, the goal is pain relief. Intra-articular corticosteroid injections are regarded as the most effective early intervention, providing rapid relief for pain and function for up to six weeks to six months.
• Physiotherapy: A combination of education and gentle exercise is often recommended. It is important that exercises in the early stages do not push through severe pain, as aggressive stretching can sometimes make the inflammation worse.
• Hydrodilatation: This involves injecting a large volume of fluid into the joint to physically stretch the tight capsule from the inside. It can improve pain and movement in the "frozen" stage.
• Surgery: If symptoms remain disabling after six months of non-surgical care, more invasive options may be considered. These include Manipulation Under Anaesthesia (MUA), where a surgeon moves the arm to stretch the capsule while you are asleep, or Arthroscopic Capsular Release (ACR), where a surgeon uses a small camera and tools to precisely cut the tight parts of the capsule.
• The UK FROST Trial: This major study found that neither Manipulation Under Anaesthesia, Arthroscopic Capsular Release, nor structured physiotherapy was clearly superior to the others at 12 months, though MUA was found to be the most cost-effective option for secondary care
Suggested Rehab plans¶
An ideal physiotherapy programme for frozen shoulder is increasingly tailored to the patient's level of "tissue irritability" (how easily your pain is triggered).* timeframes, as the clinical course varies significantly between individuals. Rehabilitation generally follows a transition from pain management to intensive restoration of motion and strength.
Phase 1: Freezing/Inflammatory (High Irritability)¶
In this stage, pain is the dominant feature, often occurring at rest and during the night.
• Goals: Reduce inflammation, educate the patient, and maintain current mobility without exacerbating symptoms.
• Education: Patients are taught to use pain as a guide and avoid activities that cause a sharp "crescendo" of pain.
• Exercises: Gentle, pain-free range of motion (ROM) is prioritised.
◦ Pendulum exercises: These use gravity to gently oscillate the joint.
◦ Active-assisted ROM: Using a cane or a pulley for gentle elevation or external rotation (ER) only within a pain-free "envelope".
• Manual Therapy: Low-grade joint mobilisations (Maitland grades I and II) may be used to alter pain signaling without stretching the capsule.
• Avoid: Aggressive stretching or "pushing through" pain, which can be counterproductive and increase the inflammatory response.
Phase 2: Frozen/Adhesive (Moderate Irritability)¶
Pain begins to settle at rest, but stiffness becomes the primary functional limitation as capsular fibrosis matures.
• Goals: Minimise further adhesions and gradually increase tissue extensibility.
• Exercises:
◦ Cane exercises: Focus on restoring external rotation, which is the pathognomonic loss in this condition.
◦ Active exercises: Performed in the scapular plane (approximately 30--45 degrees forward of the side) to minimise subacromial impingement.
◦ Neuromuscular re-education: Targeting the "shrug sign" (premature scapular elevation) to restore proper scapulohumeral rhythm.
• Manual Therapy: Transition to moderate-intensity joint glides, increasing duration as tolerated.
Phase 3: Thawing/Resolution (Low Irritability)¶
The inflammatory phase has passed, and the shoulder is stable but stiff.
• Goals: Restore full functional ROM and return the patient to high-demand activities.
• Intensive Stretching: Aggressive stretching is now the mainstay.
◦ Low-load prolonged stretching: Holding a stretch for longer durations (e.g., 30 seconds) to promote biologic collagen remodelling.
◦ Total End Range Time: Increasing the frequency and duration of time spent at the end of the available range to permanently elongate scar tissue.
• Strengthening:
◦ Rotator cuff and scapular stabilizers: Resistance-based exercises using dumbbells or elastic bands to address disuse atrophy.
◦ Kinetic chain integration: Functional movements that incorporate the trunk and legs for sports or occupation-specific tasks.
• Manual Therapy: High-grade mobilisations (Maitland grades III and IV) and end-range "holds" are applied reportedly to break through fibrotic barriers.
Emerging and Alternative Strategies
• Supervised Neglect: Evidence from some trials suggests that for certain patients, simply continuing painless daily activities and pendulum exercises yields better results than intensive therapy.
• CNS-Focused Therapy: For patients with significant fear-avoidance, mirror therapy or motor imagery can help the brain "re-map" shoulder movement without triggering a pain response.
• Post-Procedural Rehabilitation: Following an intervention like manipulation under anaesthesia (MUA) or arthroscopic release (ACR), physiotherapy must start within 24 hours to maintain the mechanical gains achieved