Introduction

Commonly known as frozen shoulder, adhesive capsulitis (AC) was first adopted by Julius Neviaser in 1945 to better reflect the underlying pathologic fibrosis accounting for the progressive loss of active and passive range of movement (ROM) [1]. The incidence is approximately 1% in the general population based on a large prospective longitudinal study of 315,308 patients over 3 years [2]. Typically, this condition affects individuals between 40 and 70 years, with women comprising nearly two-thirds of cases, and diabetes stands out as one of the most prominent risk factors [3]. Other associated conditions include thyroid disorders, cardiovascular disease, and prolonged immobilization.

Although AC is often self-limiting, recent evidence challenges this notion, showing that many patients experience persistent stiffness and functional limitations for years if not actively treated. As such, active management is essential to restore function and shorten disease duration. The natural history, previously conceptualized as the “freezing”, “frozen”, and “thawing” stages, does not consistently result in full recovery, prompting a paradigm shift toward early and targeted intervention. Pain, sleep disturbance, and disability often have a profound impact on patients’ daily functioning and psychological well-being, further emphasizing the need for timely management. Physiotherapy remains the cornerstone of management, with pharmacologic and interventional strategies, such as intra-articular steroid injections (IAIs), hydrodilatation, and suprascapular nerve blocks (SSNBs), used to reduce pain and facilitate rehabilitation. Surgical options, including manipulation under anesthesia or arthroscopic capsular release, are reserved for refractory cases.

Given the growing evidence supporting interventional procedures in improving pain relief, mobility, and overall outcomes, this narrative review summarizes current evidence on interventional procedures for AC, highlighting their indications, techniques, and comparative efficacy.

Diagnosis and natural history

The diagnosis of AC is based on clinical examination and is characterized by a gradual onset of pain and progressive loss of both active and passive ROM with at least 25% reduction in ROM in at least two planes of shoulder movement compared to the contralateral side [4]. The latter is an important criterion as it differentiates from other painful shoulder conditions that affect more active than passive movement. However, there are a few caveats in the clinical diagnosis.

First, loss of both active and passive ROM can occur in some shoulder conditions, such as severe glenohumeral joint osteoarthritis [5]. Second, the significant loss of passive ROM reflects the underlying progressive fibrotic process with capsular thickening and contracture. However, an interesting study showed that muscle stiffness or guarding could be a factor [6]. The stiffness reflects both the underlying inflammatory/pain rather than fibrotic process or psychological factors such as fear avoidance or catastrophizing thinking [7]. The understanding of this is important as it can be a barrier for physiotherapy and prompt clinicians to consider additional interventions to lower these barriers. Lastly, the interobserver reliability of applying the clinical criteria to diagnose AC is moderate [8]. There was an attempt to determine some key clinical identifiers to aid clinical diagnosis some years ago, with a Delphi consensus among different disciplines of physicians and clinicians [9]. However, it was shown later in a validation study that these clinical identifiers were not reliable for diagnosing early AC [10]. Thus, imaging plays a supportive role in the diagnosis of AC, primarily to confirm clinical suspicion and exclude other causes of shoulder pain and stiffness. Magnetic resonance imaging (MRI) provides the most specific findings, including thickening of the joint capsule and coracohumeral ligament (CHL), and decreased capsular volume. Ultrasound can also aid in diagnosis, showing thickening of the CHL, capsular thickening at the axillary recess, and hypervascularity of soft tissue within the rotator interval, reflecting fibrovascular inflammation [11].

The natural history of AC [12] is usually described in three overlapping stages: freezing, frozen, and thawing, which together give the impression of a predictable self-limiting course. The freezing stage (typically lasting 2–9 months) is marked by progressive onset of pain, accompanied by gradual loss of active and passive range of motion as synovial inflammation and capsular thickening develop. The frozen stage (4–12 months) is characterized by persistent stiffness with a plateau in pain intensity, reflecting fibrotic contracture of the capsule and reduced capsular volume. The thawing stage (lasting up to 24 months) involves gradual recovery of motion as capsular remodeling occurs and inflammation subsides. However, several longitudinal studies refuted the natural course [13, 14], and a systematic review revealed that this triphasic course is not universal [15]. Furthermore, recent evidence indicates that without supervised intervention, many patients experience residual pain or motion limitation. Nevertheless, this three-stage framework remains clinically useful for conceptualizing disease progression and tailoring treatment strategies according to the predominant pathophysiological phase.

Pathophysiology

AC starts with inflammation, which presents with pain, which can be quite intense. The inflammatory process results in the production of pro-inflammatory cytokines (including IL-1α, IL-1β, IL-6, tumor necrosis factor-α, transforming growth factor-β, and vascular endothelial growth factor), which in turn lead to enhanced fibrotic proliferation [16, 17]. In this process, there is a dysregulated collagen synthesis as well as enhanced contractility of fibroblast (myofibroblast). This process leads to capsular thickening and contracture at different stages [16–18]. As a result, there is contracture of the capsule (obliteration of the axillary pouch), thickening of the CHL, and changes in the rotator cuff interval (capsular thickness, fat obliteration). The fibrotic process affects both intra-articular and extra-articular tissue, such as the CHL [11]. As a result, there is a profound stiffness of the shoulder.

Recent literature suggests certain chemical mediators, such as transforming growth factor-β, intracellular adhesion molecules-1, are upregulated, resulting in proliferation of the fibrotic process, and such mediator is seen in chronic inflammatory states. A strong association exists between diabetes mellitus and AC, with diabetic patients exhibiting a higher prevalence, prolonged disease course, and greater resistance to treatment. The underlying mechanism is thought to involve chronic low-grade inflammation and aberrant immune-mediated fibrotic responses. Persistent hyperglycemia induces non-enzymatic glycation of collagen, leading to the formation of advanced glycation end products (AGEs) that increase cross-linking within the joint capsule, reducing elasticity and impairing remodeling. Moreover, AGEs interact with their receptors (RAGEs), promoting oxidative stress and the release of pro-inflammatory cytokines such as IL-6 and tumor necrosis factor-α, which perpetuate synovial inflammation and fibroblast proliferation. This chronic inflammatory milieu results in capsular thickening and fibrosis characteristic of AC. Recent studies further suggest that the inflammatory and immune pathways in diabetic AC differ from idiopathic cases, with enhanced macrophage and T-cell activation, providing potential targets for novel anti-fibrotic therapies [16–18].

General scheme of management

The mainstay of management is physiotherapy. Pharmacologic and intervention procedures serve to lower the barrier to make physiotherapy more effective. Rarely, some patients suffer from severe restriction of movement and require surgery, which can be manipulation under anesthesia or arthroscopic capsular release [19]. The types of intervention and when the intervention could be considered will be discussed below.

Types of interventional procedures

In general, the main target of intervention is the glenohumeral joint. However, there are different techniques and approaches.

Rotator cuff interval vs. posterior approach

Two major approaches are commonly used for glenohumeral injection: the posterior approach and the rotator cuff interval (anterior) approach. In the posterior approach, the injectate is directly injected into the posterior glenohumeral joint (Figure 1), while in the rotator cuff interval approach, the needle is inserted into the anterior glenohumeral joint (Figure 2) [25]. A recent systematic review showed that the rotator cuff interval approach is superior pain and functional outcomes compared to the posterior approach [26]. Although the injectate of both approaches ends up in the glenohumeral joint, the difference is the pattern of spread. Both the cadaver study and in vivo studies showed that the rotator cuff interval approach allows direct delivery to the glenohumeral joint and pericapsular structures involved in AC, including the CHL [27, 28]. Several high-quality systematic reviews have synthesized the evidence for interventional management of AC, yet notable methodological and clinical heterogeneity persists across studies. Poku et al. [21] demonstrated that hydrodilatation offers short-term functional improvement compared to intra-articular corticosteroid injection, but the included trials varied considerably in distension volumes, injectate composition, and outcome measures, limiting comparability. Similarly, the network meta-analysis by Liang et al. [22] suggested potential advantages of corticosteroid injection via the rotator-interval approach; however, the certainty of evidence was downgraded due to small sample sizes and inconsistent follow-up durations. In contrast, Challoumas et al. [23] provided a broader synthesis of both conservative and procedural interventions but acknowledged substantial heterogeneity and a predominance of low- to moderate-quality randomized trials. Across these reviews, long-term outcomes beyond six months remain insufficiently reported, and few studies stratified results by disease stage or diabetes status. Collectively, these limitations highlight the need for standardized outcome measures, adequately powered comparative trials, and subgroup analyses to clarify the relative and temporal effectiveness of various interventional strategies.

Display full size

Figure 1. Posterior view of the glenohumeral joint. (Figure reproduced with permission from Philip Peng Educational Series [36]).

Display full size

Figure 2. Rotator cuff interval. (Figure reproduced with permission from Philip Peng Educational Series [36]).

Conclusions

Physiotherapy remains the foundation of treatment of AC, but adjunctive interventional procedures can provide meaningful short-term pain relief and functional improvement when pain limits physiotherapy participation. Evidence supports the short-term benefits of intra-articular corticosteroid injections, with hydrodilatation and rotator cuff interval approaches offering potential advantages in mobility and functional improvement. SSNBs show comparable efficacy to intra-articular injections for pain control, though further high-quality trials are needed to clarify their optimal role.