Frozen shoulder: a narrative review of current treatment concepts and the underlying scientific evidence

Abstract

Frozen shoulder (FS) is a challenging disorder defined by persistent shoulder pain and progressively reduced joint motion. Despite its clinical significance, the underlying pathophysiology remains incompletely understood, posing challenges to optimal management. This review examines current treatment strategies, encompassing conservative approaches as well as procedural options. Evidence comparing the outcomes, complications, and recovery profiles of these techniques is critically analyzed to guide decision-making based on patient-specific factors and disease stage. Furthermore, recent advancements in understanding the molecular mechanisms of FS and the potential for novel therapeutic approaches are discussed. This comprehensive review underscores the importance of tailored treatment strategies and calls for further high-quality research to address persistent knowledge gaps in managing FS.

INTRODUCTION

Frozen shoulder (FS), also known as adhesive capsulitis of the shoulder, is a painful and disabling condition characterized by stiffness and gradual loss of shoulder mobility [1]. Originally described by Duplay [2] in 1872 as "peri-arthritis scapulohumerale", the term "frozen shoulder" was later introduced by Codman [3] in 1934, along with diagnostic and treatment guidelines. While the exact cause remains unclear, inflammation and fibrosis of the glenohumeral joint capsule are thought to play major roles [4]. This condition affects approximately 2.4 per 100,000 individuals annually, with a prevalence of less than 1% to 2% in the general population [5,6]. FS most commonly affects individuals aged 40 to 60 and is more prevalent in women and those with systemic conditions such as diabetes, dyslipidemia, or thyroid dysfunction [7-11]. A case series analysis reported that after a mean follow-up of 4.4 years (range, 2–20 years), 41% had ongoing symptoms [12]. The natural course of the disease is characterized by three sequential stages: freezing, frozen, and thawing, which occur over a duration of 2 to 3 years [7,13,14]. FS significantly impacts daily activities and quality of life, yet its pathophysiology remains poorly understood, and its clinical course is unpredictable. A recently published update reported that varies chronic inflammatory triggers initiate FS, which at the molecular level is characterized by activation of a complex cascade of cytokines, growth factors, and fibroblasts resulting in collagen deposition and minimized matrix degradation. The authors concluded that further research in FS should focus on finding specific initiating factors and identifying possible molecular targets for treatment [15].

FS treatments include a range of interventions aimed at alleviating symptoms, restoring mobility, and improving function. Non-invasive therapies are generally the first-line treatment options, and encompass exercise programs aimed at improving flexibility and strength, along with oral medications such as nonsteroidal anti-inflammatory drugs (NSAIDs) to alleviate pain and inflammation. Intra-articular injections of certain drugs, including corticosteroids, are commonly used to reduce pain and inflammation, while hyaluronic acid (HA) injections offer joint lubrication and support tissue healing [16-19]. Platelet-rich plasma (PRP) therapy has garnered attention for its potential to accelerate healing and reduce pain, and hydrodilatation—injection of saline into the joint—has shown promise in improving mobility by stretching the capsule [20,21]. Extracorporeal shockwave therapy (ESWT) is an emerging treatment that may help alleviate pain and enhance functional outcomes, although its long-term efficacy remains under investigation [22].

In cases where non-invasive treatments do not yield sufficient relief, more invasive options may be considered. Manipulation under anesthesia (MUA) involves forcibly moving the shoulder joint while the patient is under general anesthesia to break the adhesions and restore mobility. In severe cases, surgical intervention may be required, including arthroscopic capsular release (ACR), where the stiffened capsule is surgically removed or released to restore range of motion (ROM).

While a variety of non-invasive and invasive treatment options are available, no single approach has emerged as the definitive gold standard, and the efficacies of the various treatments are still being investigated. This review aims to explore the current treatment options for FS, focusing on the scientific evidence supporting various non-invasive therapies as well as invasive options, to provide a comprehensive overview that will aid in clinical decision-making and guide future research into optimizing treatment strategies for this challenging condition.

CONSERVATIVE TREATMENT

Physical therapy is the mainstay of conservative management for FS, with the ultimate goal of relieving pain, enhancing ROM, and restoring functional ability. A typical physical therapy regimen includes pain relief physical modalities, exercise and manual therapy, mirror therapy, ESWT, and laser therapy. Pain relief physical modalities include therapeutic ultrasound, interferential current therapy, transcutaneous electrical nerve stimulation (TENS), pulsed electromagnetic field (PEMF) therapy, and continuous shortwave diathermy (CSWD).

PAIN RELIEF PHYSICAL MODALITIES

Electrotherapy modalities, including therapeutic ultrasound, interferential current therapy, TENS, PEMF therapy, and CSWD, are commonly used to manage FS. These treatments aim to alleviate pain and improve function through thermal effects, electrical stimulation, and neuromodulation [23]. Therapeutic ultrasound uses sound waves to deliver energy to deep tissues. A device with a transducer transmits these waves at frequencies of 1 or 3 MHz and intensities between 0.1 and 3 W/cm². The treatment can be continuous (constant sound waves) or pulsed (intermittent sound waves) [24,25]. It is thought that this energy increases tissue temperature, thereby leading to changes in cell membrane function and cellular growth, which in turn may promote soft tissue healing and muscle relaxation [25,26]. A study comparing therapeutic ultrasound with placebo ultrasound (not turned on) in FS patients found no significant difference in overall pain relief and shoulder function at either the 2-week or 3-month follow-ups [27]. Therapeutic ultrasound might offer benefits in the treatment of FS; however, its effectiveness remains uncertain, and further research is required to validate its efficacy.

Interferential current therapy generates a low-frequency current within the body through the application and interaction of two medium-frequency electrical currents around 4,000 Hz applied from external electrodes. This interaction produces a low-frequency, amplitude-modulated current, known as a “beat frequency,” which ranges from 0 to 150 Hz and is delivered to deeper tissues [28]. The theoretical benefits of this beat frequency include pain reduction, increased circulation, and disruption of nerve conduction. A clinical trial of FS patients found similar pain and function outcomes at 4 weeks, 4 months, and 7 months follow-up between interferential current therapy combined with exercise and electropuncture with exercise [29]. However, no studies have compared interferential current therapy with a placebo control.

TENS is based on a device with electrodes that is placed on the skin over the source of pain to deliver mild electrical currents using a small, battery-powered generator [30,31]. There are two main types of TENS: conventional TENS, which uses high-frequency, low-intensity impulses to induce a tingling sensation, and acupuncture-like TENS, which uses low-frequency, high-intensity impulses to induce muscle twitching [32]. TENS is based on the gate control theory of pain, which proposes that a "gate" in the spinal cord regulates pain signals to the brain. By delivering electrical impulses, TENS may close this gate, blocking pain transmission [32-34]. Two studies have reported improved outcomes in FS patients when TENS was combined with other treatments. In one study, two groups were evaluated: one group received hot packs, TENS, and passive stretching exercises, while the other group received hot packs, TENS, and Mulligan’s technique. Mulligan’s technique combined with TENS provided better outcomes than stretching exercises and TENS in terms of ROM, functional scores, pain, and patient satisfaction [35]. Another trial compared outcomes between a group that received TENS and mobilization and another group that received TENS, mobilization, and rotator cuff muscle strengthening exercises. Greater improvements in pain and function were observed in the latter group [36]. A comparative study found that functional scores were significantly higher in the group that received a combination of therapeutic ultrasound, TENS, and home exercise compared to home exercise alone, at both 2-week and 3-month follow-ups [37]. These findings suggest that combining TENS with other therapeutic modalities may lead to improved outcomes for FS, although further studies are needed to confirm its long-term effectiveness.

PEMF therapy, which involves the application of intermittent low-frequency magnetic fields, is theorized to provide temporary pain relief by influencing tissue regeneration and cellular activity [38,39]. While PEMF has shown promise in reducing pain in other musculoskeletal conditions, its effectiveness at alleviating FS remains largely unexplored. Most studies of PEMF therapy for shoulder pain have included individuals with various shoulder conditions, making it difficult to determine its impact on FS [40]. One randomized clinical trial compared a group treated with hot packs, passive manual stretching, and pulley exercises to another group that received the same treatment plus PEMF therapy in FS patients; this study found no significant differences in pain, ROM, or shoulder function between the two groups [41]. The usefulness of PEMF for FS therefore remains inconclusive. Well-designed clinical trials are required to establish its effectiveness.

CSWD is a therapeutic modality that uses electromagnetic energy with wavelengths ranging from 3 to 30 meters and frequencies between 10 and 100 MHz to generate heat in deep tissues [42]. Unlike other forms of diathermy that use pulsed or variable energy, CSWD delivers continuous energy, allowing it to heat deep tissues and reach areas inaccessible to superficial heating modalities like hot packs [24,42]. Although CSWD has been used to treat FS, strong scientific evidence supporting its routine use is lacking. One study reported that CSWD combined with exercise led to better functional scores at 4 and 8 weeks compared to exercise alone [43]. However, another study found no significant difference in overall pain between patients treated with CSWD, hot packs, and exercise versus those who received deep friction massage and exercise over a 2-week period [44]. Given the limited research, the benefits of CSWD remain unclear, highlighting the need for more rigorous studies to determine whether CSWD alone is a valuable treatment option for FS [23]. There is little high-quality evidence concerning the effects of electrotherapy on FS to draw firm conclusions about its effectiveness, whether used alone or in combination with other treatments [23].

EXERCISE AND MANUAL THERAPY

Exercise and manual therapy are the principal nonoperative management strategies for FS, with the goals of restoring mobility, reducing pain, and improving shoulder function. Exercise therapy is generally more active, relying on the patient's participation in exercises to regain motion and strength, while manual therapy is more passive, focusing on therapist-driven, hands-on techniques to enhance mobility and alleviate pain. Exercise therapy includes stretching and strengthening exercises. Stretching exercises for FS aim to reduce joint capsule tightness, improve ROM, and alleviate pain by gradually elongating the restricted tissues. Commonly used stretches include the pendulum exercise, wall climbing, and cross-body adduction stretches, each targeting specific movements such as flexion, abduction, and internal rotation. These exercises should be performed 3 to 5 times daily, with each stretch held for 20 to 30 seconds [45]. Strengthening exercises for FS are designed to restore shoulder function, enhance muscle support around the joint, and prevent future stiffness. These exercises typically begin with pain-free isometric contractions of the rotator cuff and scapular stabilizers, progressing to low-resistance isotonic exercises. These isotonic exercises include isometric internal and external rotation, wall push-ups, and resistance band rows, which help stabilize the glenohumeral joint and scapula. Strengthening exercises are usually performed 3 to 5 times a week, with 2 to 3 sets of 10 to 15 repetitions [45]. Recently, mirror therapy has been introduced to accelerate the rehabilitation process by enhancing motor learning and increased motivation through visual feedback from a mirror [46]. Manual therapy includes graded motor imagery [47], neuromuscular exercises [48], and proprioceptive neuromuscular facilitation [49].

FS is characterized by capsular fibrosis driven by increased transforming growth factor-beta 1 (TGF-β1) [50], which promotes fibroblast activation, collagen production, and upregulation of tissue inhibitors of metalloproteinases (TIMPs) [51], leading to reduced matrix metalloprotein (MMP) activity and impaired extracellular matrix (ECM) degradation [52]. This imbalance between MMPs and TIMPs results in excessive ECM accumulation and joint stiffness. Therefore, targeting TGF-β1 signaling and restoring the MMP/TIMP balance have been explored as therapeutic strategies. Lubis and Lubis [53] suggested that MMPs, TIMPs, and TGF-β1 are associated with FS. They found that FS patients had lower MMP-1 and MMP-2 expression but higher TIMP-1, TIMP-2, and TGF-β1 expression at baseline compared to controls. After 12 weeks, intensive stretching significantly increased MMP levels, reduced TIMP levels, and improved shoulder function more effectively than supervised neglect. Active stretching enhanced recovery by modulating MMP and TIMP levels, potentially promoting ECM remodeling and reducing tissue adhesion. This underscores the value of exercise therapy as an essential treatment for FS, with an underlying scientific rationale for why intensive stretching improves outcomes. However, research comparing exercise therapy to no treatment or studies that have directly compared exercise with manual therapy are lacking. Most studies have focused on combined treatments, with no evidence supporting the superiority of one over the other [54]. A study of passive stretching with manual therapy versus supportive therapy with exercises within the pain limits (supervised neglected group) reported that the supervised neglect group showed better clinical outcomes at the 12 month follow-up with a constant score of 76.71 compared to the other group (58.97) [55]. In a study where all participants performed pendulum exercises and were randomly assigned to intraarticular steroids, mobilizations, ice therapy, or no treatment, manual therapy (mobilization) showed no advantage over the other approaches [56]. A comparative study of exercise versus oral corticosteroids reported that exercise therapy was superior at improving abduction (96.00° to 154.59°, 12th week) and external rotation (41.00° to 70.59°, 12th week), with greater effect sizes for improvement observed in the exercise group [57]. Strengthening and stretching exercises can effectively reduce stiffness and prevent the progression of FS [58,59]. Strengthening exercises such as isometric external rotation, resistance band abduction, and scapular push-ups help alleviate stiffness and halt the progression of FS by improving shoulder stability, promoting joint mobility, and reducing pain. These exercises activate key muscles like the rotator cuff and scapular stabilizers, enhance joint mechanics, increase blood flow to the capsule, and prevent further functional decline. A systematic review and meta-analysis reported that exercise therapy significantly improves ROM, function, and pain in patients with FS [59].

Physiotherapy interventions for FS can be broadly categorized into supervised and non-supervised approaches. Supervised physiotherapy involves direct guidance and monitoring by a qualified physical therapist in a clinical setting, allowing for real-time feedback, technique correction, and progressive modification of exercises. In contrast, non-supervised or home-based programs typically involve patients performing prescribed exercises independently after initial instruction, often with written or visual aids for reference. Supervised exercise sessions have been shown to provide better outcomes than unsupervised home exercises [60]. A blinded randomized controlled trial compared three conservative management strategies for primary idiopathic FS: group exercise classes, individual multimodal physiotherapy, and home exercises alone. In this trial conducted over 12 months with 75 participants, supervised group exercise classes demonstrated superior outcomes across key shoulder-specific measures than both individual therapy and unsupervised home programs [61]. However, a single-center, blinded randomized controlled trial conducted in 2017 that compared the efficacy of supervised physiotherapy combined with home exercises against a self-directed home exercise program alone in patients undergoing hydrodilatation for primary FS in 41 participants found no significant differences in clinical outcomes (Oxford shoulder score and visual analogue scale) between the two approaches at the 1-year follow-up [62]. In resistant cases where exercise protocols alone are insufficient, additional modalities can be added to improve outcomes, such as intra-articular injections [63] or hydrodilatation [64].

MIRROR THERAPY

Mirror therapy was originally developed to treat phantom pain in amputated extremities [65]. It employs a mirror to create the illusion of a functional or healthy limb by reflecting the movements of the unaffected side. This visual feedback plays a crucial role in motor rehabilitation by accelerating the rehabilitation process through enhanced motor learning and increased motivation [46]. It also promotes neuroplasticity, strengthening weakened neural pathways and benefiting conditions like stroke-induced hemiparesis [66].

Individuals adjust their posture based on visual feedback, and performing exercises with this feedback provides somatosensory stimulation while ensuring correct movement execution. In the context of FS, mirror therapy offers several advantages. By modulating pain pathways through visual and motor feedback, discomfort can be significantly reduced. Additionally, observing the mirrored movements of the healthy shoulder may stimulate motor recovery and improve ROM in the affected shoulder. Engaging the brain’s motor cortex through this therapy helps rewire neural circuits, potentially enhancing overall rehabilitation outcomes [66]. Mirror therapy is a simple, non-invasive, and low-risk intervention that patients can self-administer at home with proper guidance, making it both accessible and convenient. A prospective, randomized, controlled, single-blind study reported that mirror therapy, when combined with physical therapy, was superior to physical therapy alone in improving active flexion (P=0.001), active abduction (P=0.02), passive flexion (P=0.002), and passive abduction (P=0.02). Furthermore, the group that received mirror therapy demonstrated greater improvements in shoulder function (P=0.003) and experienced a significantly greater reduction in pain compared with the control group (P=0.007) [67]. A comparative study of three groups—one with the affected extremity behind the mirror, a visual feedback group (both upper extremities in front of the mirror), and a control group—found that exercises performed while viewing the affected extremity in the mirror (visual feedback group) were more effective than those in the other two groups [68]. However, the efficacy of mirror therapy remains under-researched. Proper instruction and supervision are essential to ensure correct technique and maximize benefits. Future clinical studies are necessary to validate the effectiveness of mirror therapy.

EXTRACORPOREAL SHOCKWAVE THERAPY

ESWT has emerged as a promising non-invasive treatment for FS, with studies reporting significant improvements in pain relief, ROM, and functional outcomes. A study without a control group reported that ESWT effectively reduced pain and enhanced shoulder function [69]. Additionally, a controlled study comparing ESWT with oral steroid therapy demonstrated its ability to improve short-term functional outcomes [70]. Saldiran et al. [71] reported that radial ESWT was effective for patients with diabetes in the short term, regardless of energy levels. Yuan et al. [72] reported that the combination of ESWT with fascial manipulation resulted in better pain relief and functional recovery than traditional local ESWT. Nambi et al. [73] found that ESWT combined with lidocaine injections led to superior clinical and magnetic resonance imaging outcomes compared to placebo. Systematic reviews and meta-analyses support the efficacy of ESWT as an adjunctive therapy, emphasizing its potential to improve outcomes when used alongside standard treatments such as physical therapy, corticosteroid injections, and manual therapy [22,74].

While these findings affirm the benefits of ESWT in managing FS, further research is needed to confirm its efficacy due to several limitations in the current evidence-base. Many studies performed to date had small sample sizes, lacked a control group, or had a retrospective design, which can lead to biased results and reduced generalizability [69,72]. Heterogeneity in patient populations, including variability in the stage of FS (freezing, frozen, or thawing) and comorbid conditions such as diabetes further complicate interpretation of the findings of these studies [71]. Moreover, most studies have focused on short-term outcomes, such as pain relief and early functional improvements, with limited data on the long-term effectiveness and recurrence rates of symptoms following ESWT [70].

There is also significant variability in the ESWT protocols that have been used across studies, including differences in energy levels (radial vs. focused), frequency, duration, and total number of treatment sessions. This inconsistency makes it difficult to draw strong, unified conclusions or establish optimal treatment parameters. Additionally, some studies have evaluated ESWT in combination with other interventions, such as fascial manipulation or lidocaine injections, making it challenging to isolate the specific effects of ESWT [72,73]. Finally, while systematic reviews and meta-analyses have reported encouraging findings, the overall quality of evidence is moderate, as these reviews were based on studies with methodological limitations [22,74]. Future research should focus on addressing these gaps by conducting high-quality randomized controlled trials with standardized protocols, larger sample sizes, and long-term follow-up to establish ESWT's efficacy and optimal application in FS management.

LASER THERAPY

Laser therapy has emerged as a promising non-invasive treatment for FS, with various studies highlighting its potential to alleviate pain, improve ROM, and enhance quality of life. The mechanism of laser therapy in FS is primarily based on photobiomodulation. When low-level laser light penetrates the affected tissues, it is absorbed by mitochondrial chromophores. This stimulation enhances mitochondrial activity, leading to increased production of adenosine triphosphate, which fuels cellular repair and regeneration processes [75]. Additionally, laser therapy exerts anti-inflammatory effects by downregulating pro-inflammatory cytokines such as TNF-α and interleukin (IL)-1β while promoting the expression of anti-inflammatory mediators like IL-10. This modulation reduces synovial inflammation and capsular thickening [76].

Low-level laser therapy (LLLT) has demonstrated efficacy, particularly in diabetic patients, by reducing inflammation and facilitating tissue repair, as shown in a randomized controlled trial [77]. However, the lack of robust clinical data and variability in treatment protocols limit the generalizability of these findings. High-intensity laser therapy (HILT) has also gained attention for its rapid analgesic and anti-inflammatory effects. A randomized controlled study found that HILT significantly reduced pain in patients with FS for up to 8 weeks, with no effect at 12 weeks and no significant differences in ROM or satisfaction, indicating short-term benefits for pain relief [78]. A double-blinded, sham-controlled randomized trial evaluated the effects of HILT on pain, disability, and quality of life in patients with FS. The study concluded that HILT reduced pain and improved quality of life, but did not significantly impact disability or shoulder function [79]. A systematic review and meta-analyses emphasized HILT’s superiority over conventional therapies in pain reduction and functional recovery [80]. However, concerns about the high cost and accessibility of HILT devices remain significant barriers to its widespread adoption. A comparative study evaluating the effectiveness of ultrasound and HILT in treating FS reported that both ultrasound and HILT were effective at managing the condition, improving pain, ROM, and overall functionality. However, the study lacked long-term follow-up, leaving uncertainty about the sustainability of the observed benefits. Ordahan et al. [81] conducted a randomized trial of 40 patients with FS, comparing the efficacy of LLLT versus HILT at the 3-week follow-up. They reported significantly higher improvements in visual analog scale and Shoulder Pain and Disability Index (SPADI) scores in the HILT group than the LLLT group, although neither therapy provided improvement in the ROM. Additionally, combination approaches, such as incorporating laser therapy with suprascapular nerve blocks or post-isometric facilitation techniques, have been shown to enhance mobility and pain relief [82,83]. Nonetheless, the heterogeneity of patient populations and the small sample sizes in most of the studies make it challenging to drawing definitive conclusions. Additionally, variations in laser protocols and patient responses necessitate further research to standardize treatment parameters and optimize clinical outcomes.

ORAL MEDICATIONS

The most widely-used oral medications for the treatment of FS are NSAIDs and oral glucocorticoids, which help manage pain and reduce inflammation. NSAIDs work by inhibiting cyclooxygenase enzymes (COX-1 and COX-2), which are involved in the production of prostaglandins. One comparative study of corticosteroid injections and NSAID treatment showed significant improvements in pain and shoulder ROM at 24 weeks in both groups with no significant differences in flexion (P=0.51), abduction (P=0.76), external rotation (P=0.12), or internal rotation (P=0.91) between groups. Another study found that a single corticosteroid injection led to significantly faster improvement in pain, motion, and function during the first 8 weeks (P<0.001) than oral NSAIDs, though no significant differences were observed at the final follow-up (12 weeks) [84,85]. Although NSAIDs are effective for pain relief, their long-term use warrants caution due to risks including gastrointestinal irritation, renal impairment, and cardiovascular complications [86]. Therefore, NSAIDs are typically used as part of a comprehensive treatment plan that includes physical therapy and other interventions for long-term management.

Oral corticosteroids are useful in the conservative treatment of FS to provide pain relief or when pain severely limits daily activities. These drugs, which are most effective during the freezing phase, reduce inflammation, pain, and stiffness by binding to glucocorticoid receptors, inhibiting the expression of pro-inflammatory cytokines and enzymes like phospholipase A2 involved in prostaglandin and leukotriene formation [87,88]. However, their role is more limited than intra-articular injections due to potential side effects, especially with prolonged or high-dose use. Common side effects include weight gain, fluid retention, hyperglycemia, and long-term use can cause osteoporosis, muscle weakness, gastric ulcers, and adrenal suppression [88-93]. For these reasons, oral corticosteroids are typically prescribed for short periods (2 to 6 weeks) at the lowest effective dose (1 mg/kg dose equivalent of prednisolone with maximum daily dose of 40 mg/kg) and used alongside physical therapy and exercise, rather than as a stand-alone treatment [87,90].

HYALURONIC ACID

HA, a component of synovial fluid that is essential for joint lubrication and chondroprotection, is widely used for knee osteoarthritis treatment with numerous studies supporting its efficacy. In the context of FS, HA acts as a viscoelastic substance that improves joint lubrication, reduces inflammation, and alleviates pain. When injected into the shoulder, HA enhances synovial fluid viscosity, leading to smoother movement and reduced friction, which decreases pain and improves ROM. HA also has anti-inflammatory effects, interacting with synovial tissue receptors to reduce pro-inflammatory cytokine release and inhibit joint capsule inflammation. Due to its anti-inflammatory effects, its use has recently expanded to FS management [94]. Several studies have reported that intraarticular HA injections control the symptoms of FS [19,95,96]. HA injections are generally well-tolerated with a low risk of significant side effects. Some patients may experience temporary pain, swelling, or stiffness at the injection site; these side-effects usually resolve within days. Rarely, allergic reactions like redness, itching, or rash occur. HA may be less effective when the joint capsule is significantly thickened and scarred [19].

STEROID INJECTIONS

Injection of a corticosteroid into the glenohumeral joint is often used to reduce inflammation and provide short-term pain relief [97]. However, this is not a long-term solution, and is usually combined with physical therapy. Intra-articular corticosteroid injection resulted in greater improvement in passive ROM both short-term and long-term in one study [97]. Another study reported that a single corticosteroid injection, administered without image guidance, resulted in faster pain relief and earlier improvement in shoulder function and motion than oral NSAIDs [85]. The optimal injection site for corticosteroids remains an area of research, with several studies reporting that subacromial and intra-articular injections are statistically equivalent. However, subacromial injection offers the added benefit of reducing the risk of corticosteroid-related adverse effects; in particular, fluctuations in serum blood glucose levels are minimized [98]. A randomized, controlled, single-blind study found no significant differences in pain, ROM, or functional scores between ultrasound-guided posterior intra-articular and anterior extra-articular injections [99]. A randomized trial that compared ultrasonography-guided corticosteroid injection with a sham injection into the coracohumeral ligament (CHL) in patients with FS reported greater improvements in pain, ROM, and function at 12 weeks in the corticosteroid group than the control group. The study concluded that targeted corticosteroid injection into the CHL improves clinical outcomes in FS [18]. However, no studies have investigated whether corticosteroid injection into the CHL is equivalent, superior, or inferior to glenohumeral or subacromial injection.

PLATELET-RICH PLASMA

PRP has been evaluated in animal models for the prevention of secondary FS, with results demonstrating a more significant reduction in structural changes in the posterior synovial membrane in the PRP group than the control group [100]. Clinical trials using both autologous and allogeneic PRP have shown its effectiveness in reducing pain and improving shoulder ROM in patients with FS [20,21,101,102]. Allogeneic PRP has been shown to reduce levels of pro-inflammatory cytokines, specifically under inflammatory conditions, indicating its safety and potential effectiveness in managing FS [101]. The regenerative effects of PRP on various tissues are well-documented and are not limited to fibrosis alone, nor appear to be explained solely by growth factors. Recent studies have shown that PRP exerts its effects through a combination of extracellular vesicles and soluble factors, contributing to macrophage polarization toward the M2 phenotype—an alternatively activated state associated with anti-inflammatory, immunomodulatory, and tissue-repair functions. M2 macrophages secrete cytokines such as IL‑10 and TGF‑β and promote ECM deposition and angiogenesis, which are essential for tissue healing and regeneration. This phenotype contrasts with the pro-inflammatory M1 state, and together, they represent the two principal polarization states of macrophages. PRP has been shown to enhance M2 polarization while suppressing M1 markers, supporting its potential role in modulating inflammation and promoting tissue repair [103,104]. PRP can trigger a rapid and strong inflammatory response, but it swiftly transitions from simply promoting fibrosis or granulation tissue to actively supporting regeneration. A key mechanism underlying this effect is PRP’s ability to influence macrophage polarization [105,106]. Further rigorous trials are needed to confirm its safety and efficacy, especially given the short follow-up times and limitations in quantity and quality of current studies.

CALCITONIN NASAL SPRAY

Calcitonin is a peptide hormone produced by thyroid parafollicular cells in mammals. It inhibits osteoclast activity, reduces bone resorption, and is used to treat several diseases such as osteoporosis, Paget’s disease, and hypercalcemia. Salmon calcitonin, which is 50 times more potent and less prone to aggregation than human calcitonin, is the preferred therapeutic, and is available as an injectable and nasal spray [107]. Calcitonin nasal spray may relieve pain in FS by modulating central pain pathways. It increases β-endorphin release and inhibits N-methyl-D-aspartate receptors and serotonin reuptake in the central nervous system, contributing to analgesic effects, particularly for neurogenic or chronic inflammatory pain. By reducing pain, it helps patients better tolerate physical therapy and shoulder mobilization exercises, indirectly promoting recovery of ROM [108]. To treat FS, calcitonin nasal spray is typically administered at a dose of 200 IU once daily for 4 weeks. Common side effects include mild nasal irritation, such as dryness or congestion, and rare cases of headache or allergic reactions.

Calcitonin gained attention after improvements in FS symptoms were observed in postmenopausal women treated for osteoporosis with it [109]. In vitro studies have shown that salmon calcitonin reduces the synthesis of TGF-β, type I collagen, and type III collagen, as well as fibroblast adhesion, all of which are key mediators of fibrosis in FS [110]. Clinical trials, including a double-blind, randomized, controlled trial, have demonstrated that intranasal calcitonin spray leads to faster improvements in shoulder pain and function than placebo in patients with FS [109].

HYDRODILATATION

Hydrodilatation, also known as distension arthrography, is a minimally invasive treatment for FS used to reduce pain and improve mobility [111]. Originally proposed by Andren and Lundberg in 1965, intra-articular hydrodilatation alleviates symptoms by expanding the joint space through the hydraulic pressure of the injected solution [112]. The volume of mixture injected for hydrodilatation ranges from 9 to 100 mL [113]. The hydrodilatation mixture typically includes corticosteroids, a local anesthetic, and normal saline, although one study employed a combination of HA and lidocaine instead [114]. The optimal combination of injectates is unknown, and there is no consensus on the ideal volume. One systematic review reported that hydrodilatation resulted in greater transient improvements in shoulder disability and passive external rotation than intra-articular corticosteroid injection. However, hydrodilatation’s long-term effects on passive external rotation remain unclear [113]. A comparative study of hydrodilatation versus MUA found statistically significant improvements in pain, disability, and external rotation in the MUA group [115]. Furthermore, whether hydrodilatation is truly a more cost-effective and convenient option than MUA remains uncertain [113]. Hydrodilatation is often combined with a rehabilitation program to maximize recovery, offering benefits such as pain reduction and improved ROM. However, there is still no consensus on a standardized rehabilitation protocol. While hydrodilatation is generally safe, reported complications include temporary discomfort and, in rare cases, infection [116-119]. Limitations of hydrodilatation studies for FS include variable study designs, small sample sizes, short follow-up periods, and inconsistent protocols for injection and imaging, hindering assessment of the long-term efficacy and reliability of this treatment.

COLLAGEN INJECTION AND ARTERIAL EMBOLIZATION

Collagen injection in FS is an emerging experimental treatment aimed at reducing capsular fibrosis and improving shoulder mobility. Two main approaches are used: injection of hydrolyzed collagen peptides, which may promote tissue regeneration and reduce inflammation, or injection of collagenase (Clostridium histolyticum), which enzymatically degrades excess collagen to alleviate stiffness. Preclinical and limited clinical studies have reported promising results in terms of pain relief and ROM improvement [120-122]. However, further large-scale, controlled trials are needed to confirm its efficacy, safety, and establish standardized treatment protocols.

Arterial embolization is a minimally invasive treatment for FS that targets the abnormal hypervascularity associated with chronic inflammation in the shoulder capsule. By selectively occluding small arterial branches, this procedure reduces inflammatory blood flow, alleviating pain and interrupting the fibrosis cycle. Clinical studies have shown significant improvements in ROM in patients resistant to conventional therapies [123,124]. While this procedure has a favorable safety profile and offers an alternative to surgery, it remains experimental, and further large-scale studies are needed to confirm its long-term efficacy and define its role in clinical practice.

INVASIVE TREATMENTS

Manipulation under Anesthesia

MUA is a procedure used to treat FS by improving ROM and reducing stiffness. MUA is often recommended when other conservative treatments, such as physical therapy, corticosteroid injections, or hydrodilatation, have failed to provide sufficient relief. Studies have shown that MUA can lead to significant improvements in pain, shoulder mobility, and function. MUA for FS begins under general anesthesia or regional nerve block to ensure complete muscle relaxation and pain control. The surgeon first evaluates passive ROM to assess stiffness before initiating manipulation. During manipulation, one hand holds the affected arm with a short lever arm and applies movement in the sequence of forward flexion, abduction, external rotation, adduction, and finally internal rotation, while the other hand stabilizes the scapula [125]. However, this procedure is associated with several complications, such as humeral fracture, glenoid rim fracture, soft tissue injury, and brachial plexus injury. Although these complications are rare when performed by experienced practitioners [126,127], an overall complication rate of 0.4% has been reported [128]. MUA is typically followed by a structured rehabilitation program to maintain the gains in motion and prevent the recurrence of stiffness. While effective for many patients, it may not be suitable for individuals with certain underlying conditions, such as osteoporosis or severe joint instability.

Capsular Release

ACR is a surgical procedure for refractory FS when conservative treatments have failed. ACR is generally performed under general anesthesia and is minimally invasive, resulting in less postoperative pain and a faster recovery than open surgery, with patients typically experiencing significant improvements in pain, ROM, and shoulder function, especially when combined with an immediate and structured rehabilitation program [129]. A prospective randomized study found that both ACR and MUA improved pain and shoulder function, but MUA was more cost-effective. Given its good outcomes, simplicity, and cost-effectiveness, MUA remains an attractive option compared to ACR for FS [125]. According to a systematic review and meta-analysis, ACR had comparable outcomes to MUA in terms of pain relief, functional and ROM improvements. However, ACR was associated with a higher risk of severe complications than MUA [130]. Notably, ACR was associated with statistically superior long-term pain relief than MUA, although the improvement did not reach the minimum clinically important difference (MCID) [130]. To clarify which FS patients would benefit most from MCID and assess its relevance across different treatments, condition-specific MCID values should be established and incorporated into clinical studies using both anchor- and distribution-based methods. Additionally, responder analysis and long-term patient-reported outcome measures follow-up are essential for meaningful comparisons of clinical effectiveness. Another meta-analysis found that ACR produced better forward flexion outcomes at 3 and 6 months than MUA. However, no significant differences in pain scores, external rotation, or the incidence of adverse events and complications were observed between the two treatments [131].

The surgical technique used to perform ACR requires careful consideration. Axillary nerve injury, particularly during inferior capsular release, may occur as the nerve lies close to the inferior capsule. To prevent this, surgeons should maintain a safe distance of at least 1 cm from the glenoid rim, avoid deep penetration with electrocautery, and consider using blunt dissection or scissors near the 6 o’clock position. Some studies have reported that isolated posterior capsular release provides reliable improvement in pain and internal rotation [132,133]. However, ACR is currently categorized into three types: complete (360°) release, anterior-inferior release, and anterior-inferior-posterior release. There is ongoing debate regarding the efficacy of these three techniques.

Several studies have reported significant short-term and long-term reductions in pain, improvements in ROM, and improvements in overall shoulder function after complete capsular release [134-137]. Ranalletta et al. [138] reported that anteroinferior capsular release provided reliable improvements in pain and ROM at least until the mid-term follow-up. Anterior-inferior-posterior release has been shown to yield favorable clinical results in several studies [134,139,140]. A systematic review and meta-analysis comparing anterior-inferior, anterior-inferior-posterior, and 360° capsular releases found that less extensive releases yielded better functional and pain outcomes. Posterior release improved early internal rotation (not sustained) and provided early, lasting flexion gains. There were no significant differences in complication rates among the three techniques [141]. This study suggested that limited capsular release is appropriate for FS that does not respond to conservative treatment [141]. However, there is no consensus on which surgical method is most appropriate for refractory FS. Some authors also recommend that the extent of release should be individualized and should be based on preoperative motion loss [142], consistent with our practices.

Subscapularis tendon release remains a controversial issue. Historically, ACR did not involve takedown of the intra-articular portion of the subscapularis. However, intra-articular subscapularis tendon release during ACR for FS to restore ROM has been advocated. Diwan and Murrell [143] compared anterior capsular release with and without intra-articular subscapularis release, and found that leaving the tendon intact resulted in inferior intraoperative ROM and a higher prevalence of recurrent stiffness (forward flexion 121°vs. 150°, abduction 114° vs. 146°, external rotation 55°vs. 68°). A biomechanical study reported that releasing the subscapularis tendon did not significantly increase translation of the humeral head above 90°, the threshold for anterior instability [144]. Additionally, a study demonstrated that ACR combined with intra-articular subscapularis tendon release yielded good clinical outcomes in treating FS without significant loss of internal rotation strength [145]. These findings collectively support the safety and favorable clinical results of intra-articular subscapularis tendon release. Key considerations include failure of 6-month conservative treatment, diabetes mellitus-associated adhesions, profound motion loss in addition to iatrogenic risks such as nerve injury, anterior instability, and internal rotation strength deficits. Future research is needed to standardize release thresholds and evaluate long-term strength outcomes [146,147].

The necessity of performing acromioplasty during ACR for FS remains uncertain and depends on the patient’s condition, underlying pathology, and coexisting disorders. Acromioplasty may help relieve pain and improve mobility in cases of subacromial impingement or acromial abnormalities. However, routine use of acromioplasty in FS without impingement offers little benefit and may increase surgical time and complication risks. Most authors report performing subacromial debridement of granulation tissue or fibrosis and removing prominent subacromial bony spurs but do not routinely perform acromioplasty [142]. Therefore, the decision to perform acromioplasty during ACR should be individualized based on preoperative imaging, physical examination, and intraoperative findings [142]. The complication rate after ACR has been reported to be 0.6%, including superficial wound infection and brachial plexopathy [148]. Postoperative care should always include early physical therapy [149].

Experimental Treatments

Several experimental approaches have been explored to prevent fibrosis, including the use of inflammatory cytokine blockers and antifibrotic agents [121,150-155]. In laboratory-based cellular research, tumor necrosis factor-stimulated gene-6 (TSG-6) inhibited the growth of primary fibroblasts from the human FS capsule by suppressing the TGF-β/Smad2 signaling pathway [156]. That study therefore proposed TSG-6 as a potential target for inhibiting the development of FS. Another investigation identified IL-17A–producing T cells in FS, which can trigger profibrotic and inflammatory responses in diseased fibroblasts through upregulation of the IL-17RA receptor, mediated by a TRAF6/nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB)–dependent signaling pathway, as therapeutic targets. Blocking IL-17A signaling—either through IKKβ inhibition or with anti–IL-17A antibodies—was shown to reduce pathological changes in affected cells. While these findings suggest that anti–IL-17A antibodies could be a potential therapeutic option for FS, their clinical efficacy remains to be established [157].

In animal models, potential treatments such as relaxin-2, tetrandrine, vitamin C, and TSG-6 have been explored. Relaxin-2 is a native antifibrotic hormone that is upregulated during pregnancy to increase tissue laxity by promoting matrix metalloproteinase production and repressing collagen production as well as the expression of TIMPs and TGF-ß1. Multiple intraarticular injections of human relaxin-2 were shown to restore ROM and eliminate capsular fibrosis in a murine model of shoulder arthrofibrosis [153]. Tetrandrine, an alkaloid extracted from the root of Stephania tetrandra, has been shown in a rat model of FS to preserve the normal reticular structure of the joint capsule during the freezing phase and to prevent disease progression by suppressing inflammation, angiogenesis, and fibrosis. However, these results have not yet been confirmed in humans [154]. Vitamin C, an antioxidant, was postulated to prevent the development of secondary FS due to its anti-inflammatory effect, and was shown to reduce thickening of the axillary recess [158].

Several drugs for the treatment of FS, including Adalimumab and collagenase C. histolyticum (CCH) have been evaluated in clinical trials. Adalimumab, a TNF-α blocker, administered subcutaneously or intra-articularly did not alleviate FS symptoms [150,151]. CCH is a combination of two collagenases (CCH-I and CCH-II) and an Food and Drug Administration-approved enzymatic injection treatment for Dupuytren’s contracture [155]. One study reported that extra-articular injection of CCH improved ROM in FS [155], while another study found functional improvement without statistical significance at the injection site or in the overall study cohort [121]. Overall, although experimental drugs have had promising effects in animal studies and early clinical trials for managing FS, further research is needed to establish their safety and efficacy in humans.

DISCUSSION

We explored the complex and enigmatic condition of FS in this review, revealing significant gaps in our understanding despite the substantial impact of this condition on patient quality of life. FS affects approximately 2.4 per 100,000 individuals annually, with a prevalence of 1%–2% in the general population, predominantly affecting women aged 40–60 years [5,6]. Patients with systemic conditions such as diabetes, dyslipidemia, or thyroid dysfunction appear more susceptible to this condition than the general population. Natural progression of this condition through freezing, frozen, and thawing stages typically spans 2–3 years, yet a significant proportion of patients (41%) experience persistent symptoms even after 4.4 years [7-12]. Recent molecular research has illuminated the underlying pathophysiological mechanisms, which involve chronic inflammatory triggers, cytokines, growth factors, fibroblast activation, collagen deposition, and reduced matrix degradation, improving our understanding of this condition [16].

Clinical Implications of Current Treatment Modalities

The current treatment landscape for FS encompasses a spectrum of interventions with varying levels of evidence supporting their efficacy. Physical therapy remains the cornerstone of conservative management, though the evidence supporting specific modalities lacks consistency. Pain relief physical modalities including therapeutic ultrasound, interferential current therapy, TENS, PEMF, and CSWD have demonstrated mixed outcomes in clinical studies. Notably, TENS combined with other therapies showed promise in improving ROM, functional scores, and reducing pain. Exercise therapy, particularly intensive stretching, has been shown to modulate MMP and TIMP activity, potentially promoting extracellular matrix remodeling and reducing tissue adhesion. This wording is scientifically more accurate and therefore preferable. However, there is insufficient evidence to establish the superiority of any single approach; most research has focused on combined interventions rather than compared the effectiveness of individual interventions. We suggest clinicians adopt individualized, multimodal approaches tailored to patient-specific factors and disease stage. MUA and ACR represent two principal surgical options for refractory FS, each with distinct advantages and limitations. MUA, performed under general anesthesia or regional nerve blockade, involves controlled manipulation of the shoulder through specific movement sequences to break adhesions and restore ROM. Studies report significant post-procedure improvements in forward flexion, abduction, and external rotation, particularly when combined with structured rehabilitation. ACR, a minimally invasive arthroscopic technique, directly addresses capsular contractures by releasing targeted portions of the glenohumeral joint capsule. Comparative analyses have indicated that both procedures yield comparable short-term functional gains, though MUA is more cost-effective than ACR due to its lower procedural complexity.

An integrated approach to FS management appears most promising based on current evidence. This approach should incorporate appropriate physical modalities for pain relief, progressive exercise therapy targeting ROM and strength, and judicious use of more invasive interventions when conservative measures fail. Such an approach should be stage-appropriate, recognizing the different therapeutic needs during the freezing, frozen, and thawing phases of the condition. Furthermore, addressing comorbidities such as diabetes and thyroid dysfunction may improve outcomes given their established association with FS pathogenesis. Clinicians should adopt shared decision-making with patients, carefully weighing potential benefits against risks while considering individual preferences, functional demands, and quality of life implications.

Future Directions for Research and Clinical Practice

Future research in FS management should prioritize several key areas to address persistent knowledge gaps. First, well-designed prospective studies comparing the efficacy of different treatment modalities, both in isolation and in combination, are needed to establish evidence-based clinical guidelines. Second, investigation into specific molecular targets identified in recent basic science research could lead to novel therapeutic approaches targeting the underlying pathophysiological mechanisms rather than merely addressing symptoms. Third, standardization of outcome measures and treatment protocols would facilitate more meaningful comparison across studies. Fourth, identification of patient-specific factors that predict response to different treatments would enable more personalized management strategies, potentially improving outcomes and reducing treatment duration. Finally, longer-term follow-up studies are essential to better understand the natural history and long-term outcomes of various interventions, particularly given the significant proportion of patients who experience persistent symptoms beyond the expected recovery timeline. Addressing these research priorities would substantially advance our understanding and management of this challenging condition.

Limitations of Current Evidence

The current evidence-base for FS management is constrained by several methodological limitations. Many prior studies have had small sample sizes, heterogeneous patient populations, and variable definitions of FS, complicating the interpretation and generalizability of their findings. Additionally, the lack of standardized outcome measures and treatment protocols hampers direct comparison among studies. The natural tendency toward spontaneous improvement in FS further complicates assessment of treatment efficacy, necessitating properly controlled trials with adequate follow-up periods. These limitations underscore the need for higher quality research employing rigorous methodologies to establish definitive treatment recommendations and clarify the relative efficacy of different approaches.

CONCLUSIONS

In many cases, FS resolves on its own over time, though the recovery process can be slow, taking from several months to a few years. There is no universally standardized physical therapy or conservative treatment protocol for FS, as treatment plans are often individualized based on the severity of the condition, the patient's specific symptoms, and their overall health. However, there are general treatment approaches that are commonly used and have shown effectiveness. Conservative treatments aim to alleviate pain and improve function while the shoulder gradually heals; patience and adherence to physical therapy are crucial during this phase. While conservative treatments are commonly used, the effectiveness of each can vary from patient to patient. Some individuals may require more intensive interventions, such as MUA, or surgical procedures, such as ACR, if conservative measures do not yield sufficient relief. Therefore, a tailored approach based on the individual’s response to treatment is essential.