Assessment and diagnosis of non-traumatic shoulder instability: A scoping review

Abstract

Background

Given its complexity, there is no consensus regarding the assessment of non-traumatic shoulder instability (NTSI) to this day. We, therefore, conducted a scoping review to map the existing white and grey literature regarding diagnostic and assessment tools for the NTSI population.

Method

We followed the Arksey and O'Malley five-stage guideline for the conduct of scoping reviews and searched through 12 electronic databases for English-language articles and reviews from 2000 to 2024 related to NTSI's diagnosis and assessment tools.

Results

Among the 3426 identified studies, 58 were included (describing 59 different interventions). Case–control studies (43.1%, 25/58) and narrative reviews (34.5%, 20/58) were the most prevalent. Diagnostic imaging was the most studied intervention (35.6%, 21/59). Twenty-seven studies specified a direction of instability, of which 59% (16/27) were multidirectional instability. Non-traumatic shoulder instability often affects young individuals, with complex symptoms, including neuromotor deficits, muscular imbalances and kinematic alterations, involving psycho-behavioural and somatosensory components.

Discussion

Non-traumatic shoulder instability's aetiologies and clinical manifestations are multifactorial. The prevalence and incidence of this dysfunction are probably underestimated. Clinical history is crucial to retrace a complex and chronic dysfunction. The use of orthopaedic shoulder tests and the routine use of imaging currently appear to have limited relevance as a first-line approach.

Introduction

Non-traumatic shoulder instability (NTSI) is defined as abnormal movement or position of the shoulder, leading to pain, subluxation or even dislocation and functional discomfort without a significant history of injury. ¹

Since the 1980s, 18 classifications ² have been developed in an attempt to define the sub-groups of instability, but none of them have demonstrated strong measurement properties. ³ These classifications aim to determine a prognosis and define the most appropriate treatment.

In 1979, Rockwood ⁴ distinguished shoulder instabilities based on the presence or the absence of trauma, although the definition of a traumatic etiology was a source of debate. The Thomas and Matsen ⁵ classification followed Rockwood's dichotomy between traumatic and non-traumatic dislocations. They introduced the AMBRI (atraumatic multidirectional instability [MDI], bilateral, often requiring rehabilitation and inferior capsular shift) versus TUBS (traumatic unidirectional instability with Bankart lesion, often requiring surgery) acronyms. This proposed dichotomy did not seem to match the presentation of some patients met in the clinic, especially those presenting with muscle patterning problems and/or voluntary instability. Thus, the Stanmore classification ⁶ was introduced in 2004, encompassing all three categories: the Polar type I (true TUBS, traumatic, structural), the Polar type II (true AMBRI, atraumatic, structural) and the Polar type III (atraumatic with functional origin due to muscle patterning disorders or habitual non-structural), as well as all the presentations of instability that combine several polar types as a dynamic continuum.

Among all existing classifications, the FEDS classification, ⁷ introduced in 2010, is the only one that has undergone reliability testing. It categorizes shoulder instabilities based on four key characteristics: frequency (solitary, occasional or frequent); etiology (traumatic or atraumatic); direction (anterior, posterior or inferior); and severity (subluxation or dislocation). However, a recent scoping review revealed that this classification has been scarcely used in rehabilitation science research since its publication, with only 4% and 8% of studies referencing the Thomas and Matsen and Stanmore classifications, respectively. ⁸ Notably, the distinction between traumatic and atraumatic onset of primary instability appears to be a key component in evaluating shoulder instability.

A recent RCT also asserts the specificity of the management of non-traumatic dislocations, given the lack of efficacy of the surgical (arthroscopic capsular shift) treatment compared to a sham surgery. ⁹ A conservative approach including exercise is indicated as the first-line treatment for NTSI^10–12.

As exercise prescription is a key component of management, the clinical assessment should assess motor strategies that may contribute to functional limitations and associated pain, as well as social and psychological factors, to provide person-centered care. We conducted a preliminary mapping review ¹³ on the subject to guide the commissioning of a more in-depth analysis. This initial review revealed a significant lack of literature on the topic and underscored the pressing need to systematically identify the tools available to clinicians for assessing NTSI.

Consequently, the aim of this scoping review is to systematically map the existing literature on the diagnostic and assessment tools available to clinicians for managing NTSI.

Methods

We conducted this review in accordance with the Arksey and O’Malley Scoping Review methodology. ¹⁴ All relevant studies meeting our selection criteria were included, irrespective of their quality. ¹⁵

To ensure comprehensiveness, we collaborated with an information specialist to identify several databases, including sources of peer-reviewed and grey literature. Search strategies were tailored to each database. The results were then extracted and selected using Rayyan software. ¹⁶ Detailed descriptions of each step are provided in Supplementary material.

The scoping review protocol was registered on Open Science Framework: osf.io/4nd25

Eligibility criteria

Table 1 presents the eligibility criteria. Studies involving mixed populations (both traumatic and non-traumatic) were included in the title and abstract selection phases and were retained in the full-text selection phase if they reported results specific to non-traumatic instability. Both adult and adolescent populations (12-year-old and above) were included because of the higher prevalence of NTSI in adolescents (i.e., 16-year-old). ¹⁷

Table 1.

Eligibility criteria.

Criterion	Inclusion	Exclusion
Time period	Studies from 2000 to 2024
Language	English
Type of article	Reviews or articles published in peer-reviewed or non-peer-reviewed journals, master theses, doctoral thesis, conference proceedings, books or book chapters
Study topic	Diagnosis and/or assessment of non-traumatic shoulder instability	Bone loss population Post-operative population Management study
Population	Human Adults Adolescents (12-year-old and above) Non-traumatic shoulder instability	Animal Cadaveric Adolescents (under 12-year-old) Traumatic shoulder instability

Information sources

Studies were extracted in two steps. Initially, we extracted articles from peer-reviewed and grey literature. Given the limited number of articles on NTSI found in the initial peer-reviewed search, we expanded our scope to include reviews in the grey literature search and conducted a complementary search that incorporated reviews from the peer-reviewed literature. To ensure thoroughness, we also performed a snowball search during this supplementary phase.

We searched the following databases: PubMed, Embase, Cochrane, ScienceDirect, SPORTDiscus and PEDro.

Grey literature searches were conducted using Google Scholar (following Godin's ¹⁸ search method), BASE, WordWideScience, OATD and EThOS. Additionally, we searched for studies and guidelines from the websites of English-language shoulder-learned societies.

The complementary search for reviews was conducted in PubMed, Embase and Cochrane.

We included publications from 2000 to 2022, and an update March 2024 yielded no new inclusions. Detailed combinations of keywords, databases and search strategies are provided in Supplemental material.

The snowball search involved examining the references of included studies to identify additional relevant studies not captured in the initial database search.

Selection of sources of evidence

All the records identified in the initial peer-review article search were collated and uploaded into the Rayyan software, ¹⁶ and duplicates were removed. Three reviewers independently screened the studies by title, abstract and full text against the selection criteria.

For the grey literature, all identified records were similarly collated and duplicates were removed. The selection process was conducted by two reviewers, following the same procedure as for the peer-reviewed literature. Studies from English-language Shoulder Society websites were also independently screened by two reviewers and added to the list of grey literature studies.

Any disagreements during the selection process were discussed and resolved before proceeding to the next phase. The complementary search was conducted in the same manner as the previous searches.

The complete search results are reported in a PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) flow diagram (Figure 1).

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) flow diagram.

Data extraction and synthesis

A data extraction table was developed a priori to consistently collect information on the following items: (1) author; (2) year of publication; (3) country; (4) type of evidence source; (5) methodology; (6) study design; (7) population; (8) direction of instability; (9) sub-groups; (10) interventions, criteria or tools used for diagnosis; and (11) results. This framework guided the data extraction process. Two reviewers independently extracted the data, and any discrepancies were discussed and resolved.

Based on the collected data, a thematic narrative synthesis was performed, organizing the information into four main categories, each with relevant sub-themes: clinical history, physical examination, orthopaedic shoulder tests (OSTs) and imaging.

We did not assess the risk of bias for the included studies, in accordance with scoping review methodological guidelines. ¹⁵

Results

Selection of sources of evidence

The search yielded 3426 results, and after removing duplicates, 3005 records were screened by title and abstract. Subsequently, 238 full-text articles were reviewed, and a total of 58 publications met the eligibility criteria for final inclusion. The selection process, along with the reasons for excluding full-text papers (both peer-reviewed and grey literature), is detailed in Figure 1.

The authors reached consensus during the selection process, so no third party was required to resolve disagreements. The most common reason for exclusion was the inclusion of a traumatic population or a mixed population (i.e., both traumatic and non-traumatic) without specific analyses of non-traumatic sub-groups.

Characteristics of sources of evidence

Table 2 provides a summary of the characteristics of the 58 included studies. Howard's thesis ¹⁹ included two relevant interventions, bringing the total number of interventions to 59.

Table 2.

Characteristics of the included articles.

Article characteristics	n (%)		Article characteristics	n (%)
Total	58 (100)		Type of instability
Country of origin*			Unspecified	31 (53.4)
Europe	35 (60.3)		Multidirectional (MDI)	16 (27.6)
North America	11 (19)		Posterior	6 (10.3)
Asia	10 (17.2)		Anterior	5 (8.6)
Australia	2 (3.4)		Studied intervention**
Type of evidence source			Diagnosing imaging	21 (35.6)
Clinical trial	31 (53.4)		General review of shoulder instability exam	18 (30.5)
Review	23 (40)
Thesis	3 (5.2)		Electromyogram	5 (8.5)
Discussion	1 (1.7)		Clinical test	3 (5.1)
Methodology			Arthroscopy	3 (5.1)
Quantitative	30 (51.7)		Orthopaedic shoulder test	3 (5.1)
Qualitative	23 (40)		Classification	2 (3.4)
Mixed methods	5 (8.6)		Functional imaging	2 (3.4)
Study design			History	2 (3.4)
Case–control	25 (43.1)		Motion capture	1 (1.7)
Narrative review	20 (34.5)		Physical examination	1 (1.7)
Cross-sectional	9 (15.5)		Tissue analysis	1 (1.7)
Case report or series	1 (1.7)
Systematic review	1 (1.7)
Guidelines	1 (1.7)
Delphi	1 (1.7)

*Asian countries included: China, Japan, Korea, Saudi Arabia and Taiwan; European countries included: Austria, Belgium, Denmark, France, Germany, Ireland, Italy, Spain, Switzerland and UK; North American countries included: Canada and USA.

**Howard (2016) thesis comprised two studies which increases to 59 the number of studies.

The studies were published between 2000 and 2022, with half published before 2013.

The 31 clinical trials included control groups,^20–50which were composed of healthy volunteers in 17 studies,^{20,22–26,28,30–33,35,38–40,44,48} patients with traumatic instability in 7 studies,^{29–31,33,34,39,45} and individuals with stable shoulders but other shoulder disorders such as rotator cuff tears or pathologies, in 5 studies.^{21,25,27,30,37} Six studies had heterogeneous comparison groups with several types of controls.^{25,30,31,33,37,39}

The NTSI population across all included studies had a mean age of 25.71 years [range: 18–37.8], a median age of 24.5 years [Q25 = 22; Q75 = 28], and a standard deviation of 4.24 years. The gender distribution across the studies, cumulatively including both the investigated and comparison groups, comprised 1227 men and 1030 women, resulting in a slightly higher male-to-female ratio (1:19).

Results of individual sources of evidence

Clinical history

The patient's history and physical examination are crucial components of the clinical assessment. Various risk factors contribute to NTSI, which primarily affects individuals aged 15–35, with a higher prevalence in those under 25.^1,47,51–54 The clinical examination process is similar for both young adolescents and adults. ¹⁷ Most NTSI patients experience recurrent subluxations (more than three per year ⁵⁵ ) rather than true dislocations.^54–57 It is essential to determine whether the instability is isolated or recurrent, ⁵⁷ as well as to assess the shoulder's irritability ⁵⁴ and any associated muscular weakness following these events. ⁵²

Disease history

The first episode of instability usually occurs in childhood,^17,43,47,57 with severity potentially leading to chronic NTSI and increasing the risk of recurrence over time.^51,57 Chronic NTSI can result from an initial traumatic incident, particularly when the first episode occurs during childhood, which exposes individuals to a higher recurrence risk. ¹⁷ It is crucial to adopt a holistic approach that considers the patient's entire history, experiences of instability, reactions and body awareness, rather than focusing solely on the most recent episode or the shoulder joint itself. ⁵⁸

Patient description of symptoms

During the patient's clinical history, the severity of the patient's symptoms must be thoroughly investigated, ranging from minor disruptions in sports activities to significant functional impairments in daily tasks. ¹⁷ Symptoms can vary widely ⁵¹ and may be unclear,^51,59 with patients often reporting a subjective sense of instability, ⁵² as well as clicking, catching, cracking, popping or grinding sensations, sometimes accompanied by pain.^17,57,59 They may also experience increased tenderness in the posterior glenohumeral joint (GHJ) ⁵² or pain over the long head of the biceps tendon, ¹ which can radiate down the arm. It is important to assess any lifestyle changes or reduced sports performance, particularly in overhead activities. ⁵³

Voluntary component of instability

During the clinical examination, the clinician must determine whether the instability has a voluntary component.^{1,47,56,59–61} Moroder et al. ⁴⁷ describe functional shoulder instability (FSI) as a sub-group of NTSI characterized by abnormal muscle activation without structural changes. Functional shoulder instabilities can be positional (occurring during arm movement) or non-positional (at rest, via muscle contraction), with 78% being positional. ⁴⁷ Assessing the patient's control over instability episodes is crucial, as less control leads to greater functional repercussions. ⁴⁷ Moroder et al. ⁴⁷ also distinguish between voluntary instabilities associated with psychiatric issues or secondary gains^47,59–61 and ‘demonstrable’ instability (e.g., during festive events),^1,47,60,61 where patients can dislocate their shoulder voluntarily but do not do so regularly. Identifying FSIs is essential for guiding treatment toward a global, multidisciplinary approach, as these patients are prone to surgical failures.^59,60

Sleeping disturbance

Non-traumatic shoulder instability can disrupt sleep^42,58,61 by causing pain, inability to lie on the affected shoulder or night-time instability, leading to awakenings that may exacerbate symptoms and contribute to psycho-behavioural issues. ⁵⁹

Psycho-behavioural impact

Lebe et al. ⁴² highlighted the role of psycho-behavioural factors in worsening symptoms in patients with recurrent NTSI, with 40% of their cohort exhibiting significant depression, self-harm or suicide attempts. This group often reported higher levels of pain, anxiety, sleep disorders and poor dominant arm function compared to healthy individuals.^42,58,61 The severity of these issues correlated with both functional impairment and symptom duration. ⁴² However, no severe mental health disorders were identified. Moroder et al. ⁴⁷ found that patients with uncontrollable instability, particularly in the non-positional group, had worse scores, while those with controllable instability, who did not perceive their condition as pathological, had better outcomes. Tools such as the Fear-Avoidance Beliefs Questionnaire, the Hospital Anxiety and Depression Scale, the Pain Catastrophizing Scale and the Personal Self-Efficacy Score may be useful for exploring psycho-behavioural aspects and identifying triggers of conditioned responses. ⁶¹

Somato-sensory modification

Non-traumatic shoulder instability affects somatosensory perception, altering sensory-motor cortical representations and body perception. ⁶¹ Repeated episodes of instability often lead patients to avoid unstable movements or positions, resulting in apprehension or kinesiophobia.^1,53,61 Functional magnetic resonance imaging (MRI) studies show that this apprehension reorganizes brain areas related to anxiety, emotional regulation, motor control and pain perception, as well as pathological brain plasticity linked to the over-connection of abnormal motor pathways. ⁶² These changes lead to increased brain activation, particularly during simple motor tasks, contributing to overall instability.^19,38,60 The Rowe and WOSI scores are used to assess the functional impact of NTSI, with the Rowe score being more comprehensive as it incorporates aspects of apprehension through its range-of-motion (ROM) items. ⁶²

Physical examination

The physical examination, conducted bilaterally,^{26,45,57,58,60,63} aims to exclude other causes of shoulder pain, such as rotator cuff tendinopathy or frozen shoulder,^56,59 and to identify coping strategies that can be addressed during symptom modification procedures, forming the basis for subsequent treatment. Instability in NTSI results from a combination of dysfunctions rather than a single deficit. ¹⁹ NTSI is a complex syndrome involving capsuloligamentous laxity, impaired muscle function, neuromotor control issues and osteoarticular abnormalities.^27,60 Initial morphostatic observation can provide important clues, such as muscular atrophy, abnormal shoulder posture and overall body posture.

Apprehension and symptom provocation manoeuvers

During the clinical history, patients with NTSI may report apprehension or insecurity about their shoulder and may use coping strategies to avoid certain positions or movements.^1,53,61 Anterior instability typically occurs in the ABER position (abduction and external rotation [ER]), which is common in arm-reaching movements such as throwing.^59,64 Posterior instability often arises with 90° flexion, adduction, internal rotation (IR) and a posterior force, as seen in activities such as push-ups, pushing heavy doors, bench pressing or football blocking.^51,55,59,65 This type of instability is most frequent in individuals classified as Polar Type III^47,60 (78%, according to Moroder et al. ⁴⁷ ). Inferior instability can be induced by carrying heavy loads,^17,51,59 causing inferior traction on the humeral head, which may be highlighted by an exaggerated sulcus sign (>2 cm) or peripheral neuropathy due to brachial plexus traction. Non-traumatic shoulder instability may also involve MDI. ⁵⁶

Patients might experience apprehension even without new instability episodes and with a shoulder that appears clinically stable, which can be explained by micromovements in the GHJ, sometimes involving peripheral neuropathy. ⁶² Clinicians may need to challenge patients to identify these micro-instabilities.

Hyperlaxity

Risk factors for NTSI include participation in specific activities or sports, particularly performance sports, weightlifting, throwing and any overhead activities, whether in a sporting or professional context. These activities increase the risk of musculoskeletal injury due to repetitive strain and overuse of the shoulder.^{17,43,46,47,52–55,57,59,60,64–66} Such repetitive stress can lead to acquired joint hyperlaxity, a common symptom in NTSI patients,^{1,26,43,51,53,54,58–61} especially among women,^51,56 who may also report lifelong laxity. ⁵⁴ Hyperlaxity can be acquired through these activities or be congenital, as seen in conditions such as congenital hypermobility syndrome, Marfan syndrome or Ehlers–Danlos syndrome hypermobile type.^{1,17,58,59,61,63,65}

Milewski et al. ¹⁷ suggest that all young individuals with suspected shoulder instability, as well as adults with recurrent instability, should be assessed for generalized hyperlaxity, including an evaluation of family history and other symptomatic joints (e.g., recurrent ankle sprains).^17,58,59,61 Castagna et al. ³⁹ found that NTSI patients, particularly younger individuals, exhibit a greater density of elastic fibres compared to those with traumatic instability and healthy populations. ³⁹ This hyperlaxity tends to decrease with skeletal maturation and age. ⁵⁶

Many authors caution against directly linking hyperlaxity with instability,^56,59,65 as excessive GHJ translation does not necessarily imply instability.^59,65 Hyperlaxity is considered a risk factor but not a direct cause of NTSI. According to Warby et al. ⁵¹ and the Stanmore classification, ⁶ NTSI patients may evolve from congenital to acquired hyperlaxity and from a non-traumatic to a traumatic etiology, with joint laxity potentially affecting proprioception and sensorimotor control. ⁵¹ However, the exact impact of joint laxity on proprioception remains unclear.^20,51,58,60

Articular range of motion

When assessing ROM in atraumatic hyperlax instabilities (Polar Type II), specific combinations, such as excessive ER, a positive sulcus sign and internal rotation deficit, may be observed.^52,60 A deficit of more than 25° in IR of the throwing arm compared to the opposite side is considered clinically significant. ⁶⁰ It is essential to thoroughly evaluate the full range of motion of the GHJ, including its outer range function. ⁵⁸ Assessing both active and passive GHJ rotations at 90° of abduction can help detect scapular instability, ⁶⁰ and reduced scapular rotation during arm elevation may also be present. ⁶⁷

Muscular function

Electromyographic (EMG) studies have revealed abnormal muscle activation patterns with altered frequency and latency in muscles^20,40,56,61 such as the Pectoralis Major (PM), Pectoralis Minor (Pm), Latissimus Dorsi (LD), Deltoid (DT), Subscapularis (SSC), Infraspinatus (IS), Supraspinatus (SSP), Trapezius (Tr) and Rhomboids.^40,58,60 However, no single muscle dysfunction is solely responsible for instability ¹⁹ ; these patterns often reflect adaptive force generation mechanisms. Muscle imbalances, such as hypertonic internal rotators and hypotonic external rotators, are frequently observed,^53,58,60,61 highlighted by an altered IR/ER strength ratio and visually indicated by a protective curling posture, commonly seen in anterior chain hypertonicity and posterior chain hypotonicity.^53,61 Moroder et al. ⁴⁷ noted hypoactivity of muscles in cases of positional instability and, conversely, hyperactivity of larger muscles in non-positional instability cases.

During physical examination, palpation can help identify muscle tone imbalances, ⁶⁰ while dynamometry is used to assess strength discrepancies. ⁵⁸ Grip strength is correlated with shoulder strength, ⁴⁸ emphasizing the importance of shoulder stability for overall upper limb function. The grip test serves as a quick and indirect method to assess shoulder strength. ⁴⁸ Muscle alterations contribute to shoulder deconditioning,^1,58,60,61 leading to strength loss,^51,53,61 increased fatigue,^58,60,61,65 and reduced dynamic stability.^58,60,61,65

Kinematics

In response to muscle alterations, patients with NTSI often modify their kinematics, exhibiting intensified and prolonged activation of stabilizing muscles around the humeral head to enhance control and stability through increased compressive forces.^61,67 Conversely, muscles responsible for arm movement and acceleration show reduced and shortened activity, while those involved in eccentric deceleration or movement control experience increased activation. ⁶⁷ Despite these adaptations, NTSI patients often demonstrate greater humeral head translation across various ROM compared to those with traumatic instability.^{26,28,45,51,53} These adjustments result in reduced shoulder velocity, which can be measured using an accelerometer. ⁵²

Patients capable of voluntary shoulder dislocation may consciously or unconsciously modulate muscle contraction or relaxation. ⁵⁹ In Polar Type III individuals, there's notable overactivation of large muscles alongside suppression of the rotator cuff. ⁶⁰ They often adopt an arm-lifting pattern involving IR with overactivation of muscles such as the Latissimus Dorsi (LD) and Pectoralis Major (PM), while inhibiting others such as the Infraspinatus (IS), lower trapezius (LTr), serratus anterior and posterior deltoid. ⁶⁰

Jaggi et al. ⁴⁹ used dynamic EMG (DEMG) to assess muscle activity in anterior NTSI, revealing that overactivation of PM and LD contributes to anterior instability. ⁴⁹ Botulinum toxin injections into the PM have been shown to facilitate the relocation of the GHJ in cases of persistent anterior dislocations. ⁴⁹ In posterior NTSI, LD overactivation causes scapular depression and excessive GHJ IR, leading to posterior translation. In MDI, LD plays a significant role in both anterior and posterior instability and inferior translation of the humeral head. The IS, crucial for GHJ compression, is often under-activated, exacerbating instability. Managing these aberrant muscle patterns poses significant challenges due to the complex dynamics of muscle activation and inhibition. ⁶¹

Electromyography

Electromyography, particularly DEMG, can be a useful tool for assessing muscle patterns in NTSI. ⁵⁰ It helps identify muscle activation patterns that contribute to instability in various directions, guiding targeted rehabilitation strategies to avoid reinforcing inappropriate muscle recruitment. ⁴⁹ Electromyographic can also serve as a tertiary diagnostic tool, aiding in understanding treatment failures or predicting surgical risks.^50,59,60 Fine wire EMG is considered the most effective modality, offering detailed insights compared to surface EMG, which lacks the sensitivity needed to detect abnormal patterns in muscles such as the PM, LD, and Anterior Deltoid, especially in mixed Polar Type II and III populations. ⁵⁰

Neuromotor control and posture

Neuromotor control is a crucial component of shoulder stability, alongside muscle function and passive stabilizers. Childhood is a key period for developing neuromotor control, with early stages, such as learning to walk, influencing sensory-motor integration and coordination, which can lead to disorders in adolescence or adulthood. ⁶¹

Signs such as lumbar hyperlordosis, thoracic hyperkyphosis or excessive scapular protraction may indicate underlying neuromotor deficits.^47,58,60 Patients may adopt a sway-back posture with hyperextended hips and knees, shifting load onto superficial muscles like the LD and causing secondary thoracic kyphosis, which results in downward and forward tilting of the glenoid cavities. ⁶⁰

During simple motor tasks, these patients often exhibit heightened cortical activation.^19,38 Challenging them with dual-task exercises (e.g., single-leg balance, kneeling on all fours or snow angels) can reveal deficits in coordination and postural control.^60,61 Physical examination should include an assessment of spinal movements (cervical, thoracic or lumbar)^58,60,61 and scapular positioning abnormalities during motion, such as detachment from the rib cage, which may indicate neuromotor issues.^51,58,60,61

Independent assessment of scapular control involves evaluating movements such as shoulder shrugging, protraction/retraction and adduction/abduction for quality, smoothness, pain or the need to support arm weight. ⁵⁸ These assessments can provide valuable insights into neuromotor deficiencies that affect shoulder stability and help guide targeted rehabilitation strategies.

Symptom modification procedure

Once the underlying mechanisms and patient coping strategies are identified, symptom-modification strategies can be used to guide treatment. Jaggi and Alexander ⁵⁸ propose various approaches targeting muscle recruitment, such as preactivating the rotator cuff, particularly the posterior cuff, with resistance during arm elevation to enhance stability. ⁵⁸ Correcting scapular positioning and reducing stress (e.g., supporting the scapula during elevation) are essential for optimizing shoulder force generation. Addressing overall posture, such as maintaining an upright stance with retracted scapulae, prepares the shoulder for optimal function. Additionally, engaging the kinetic chain – through lower limb activation, thoracic rotation and balance exercises – enhances cortical motor activation, providing additional support for shoulder function. These strategies aim to alleviate symptoms and effectively improve shoulder stability. ⁵⁸

Neurological examination and associated disorders

Patients with shoulder issues may also experience cervical spine pain, ⁵¹ which can contribute to thoracic outlet syndrome, characterized by neurological and vascular symptoms. ⁵¹ Inferior shoulder instability may lead to arm pain due to brachial plexus traction, resulting in sensations such as pain, numbness or tingling, especially during heavy lifting, potentially causing difficulties with holding or manipulating objects.^17,51,59 A neurological examination is crucial for identifying referred pain from other areas.^58,59,61 Additionally, assessing nerve mechanosensitivity through palpation and upper limb nerve tension testing can help detect nerve damage. ⁶¹ Investigating central sensitization is also important, as it significantly influences pain perception in these cases. ⁶¹

Orthopaedic shoulder test

Figure 2 summarizes the various OSTs discussed in the reviewed studies, categorized into laxity tests, provocation tests and labral lesion tests. Laxity tests were mentioned in 15 studies,^{1,32,47,51,53,54,57–61,64–66,68} provocation tests in 13 studies,^{32,51,54,55,57–61,65,66,68} and labral lesion tests in six studies.^{36,55,57,65,66,68} For provocation tests, the anterior apprehension test, relocation test and release test were frequently cited for assessing anterior instability, while the Jerk test and Kim test were most mentioned for posterior instability.

Figure 2.

Orthopaedic shoulder test (OST).

Among the studies discussing these tests, only three provided metrological data,^57,66,68 showing good values for tests such as the anterior apprehension and relocation tests. Combining these tests enhanced specificity. Many tests, such as the Dynamic Rotary Stability Test ⁶⁰ and the Corkscrew test,^58,60 assess shoulder function and neuromotor control involving lower limb and trunk stability.

However, several tests lack evaluation for clinical utility and intertester reliability.^55,66,68 Laxity tests often detect excessive humeral head translation, but hyperlaxity does not always correlate with instability. ⁵⁹ Variations in applied force among operators can affect reliability. ³² Overall, there is no gold-standard test for diagnosing shoulder instability or structural lesions,^51,56,60 underscoring the ongoing challenges in clinical assessment.

Imaging examination

Imaging serves as a therapeutic decision-making tool, helping to rule out intra-articular pathology that might require surgery or affect the outcomes of conservative treatment.^59,61,63,69 There are no specific imaging findings for diagnosing NTSI.^63,70,71 The primary goal of imaging is to detect lesions or structural changes that may be related to NTSI. In cases of non-traumatic painless voluntary instability, imaging is often unnecessary due to the absence of structural lesions that would influence therapeutic decisions. Additionally, individuals classified as Polar Type III may lack imaging findings that biomechanically explain the severity of their instability. ⁴⁷

Morpho-anatomical risk factors

Certain GHJ structural morphologies are associated with an increased risk of NTSI.^47,51,53 A flatter^{22,26,28,33,47} or abnormally retroverted glenoid,^{22,26,28,33,34,37,47,65} as well as GHJ dysplasia or bone loss ⁶⁵ are particularly associated with posterior NTSI^34,37 and result in a lower Bone Stability Ratio in NTSI patients (−40%) compared to those with traumatic instability (−20%) and healthy individuals. ³³ Diagnostic signs such as the crescent and triangle signs in the ABER (abduction and ER) position demonstrate high specificity but low sensitivity for detecting capsular laxity.^24,53 Capsular laxity, especially with inferior capsule redundancy or rotator interval insufficiency, is often linked to MDI.^25,51,53 Although capsular elongation is common in NTSI without significant structural changes,^{53,59,63,69,71,72} the lack of standardized methods for measuring capsular redundancy limits its clinical impact on treatment decisions.^24,69 Examination Under Anaesthesia remains the gold standard for assessing joint laxity.^57,59,66 Interestingly, humeral head morphology does not significantly differ between atraumatic, traumatic or healthy groups, suggesting that other factors contribute more substantially to NTSI risk. ⁴⁷

Structural lesions

Non-traumatic shoulder instability may involve structural lesions,^29,41 such as bone injuries (e.g., Hill-Sachs) or ligament injuries (e.g., middle glenohumeral ligament).^{29,41,46,47,53,59,63,71} However, these injuries tend to be less severe than those seen in traumatic instability due to the lower kinetic forces involved.^29,41,56 Labral and rotator cuff lesions are more prevalent in NTSI, often resulting from repeated microtrauma, particularly in individuals engaged in overhead sports. ⁶⁶ Despite these structural issues, imaging does not reveal significant differences between the shoulders of adolescents with NTSI and those of healthy adults ¹⁷

Imaging modalities

There are three primary imaging modalities used to diagnose structural alterations in NTSI: radiography, computed tomography (CT) and MRI, especially when conducted with arthrography (MRA). Ultrasound is not commonly employed in diagnosing NTSI-related lesions but is instead used to assess associated rotator cuff pathology. ⁷²

Radiography is the first-line modality for excluding dislocations and assessing bone or cartilage abnormalities.^53,59,72 Various views, such as the anteroposterior (AP) view, true AP view (Grashey view), axillary view and modified scapular Y view (outlet view), are typically utilized.^{17,53,59,72,73}

Computed tomography offers more detailed bone assessment than radiography, allowing for quantification of bone loss and evaluation of morpho-anatomical parameters, although it involves higher radiation exposure.^59,65,73 Magnetic resonance imaging, often preferred for cases of minor instability or when radiography is normal, provides superior soft-tissue visualization, making it particularly useful for detecting labral, ligament or capsular lesions that may indicate a need for surgery.^47,53,59,73 Magnetic resonance imaging with arthrography enhances contrast and sensitivity for assessing joint structures^17,53,72 and is especially effective in the ABER position.^24,45,46,69 Computed tomography arthrography, though invasive, serves as a complementary tool for soft-tissue imaging when MRI is not feasible. ⁷²

Discussion

The current literature on NTSI reveals both quantitative and qualitative limitations, primarily due to the overwhelming focus on traumatic shoulder instability. This focus partly explains the scarcity of dedicated studies on atraumatic forms of instability. Non-traumatic shoulder instability often manifests as a complex syndrome driven by a combination of anatomical predispositions, abnormal muscle function and deficits in neuromotor control, rather than being a straightforward nosologic entity.

Clinical assessment and population insights

Diagnosis of NTSI heavily relies on clinical history and physical examination.^{17,51,55,56,58–60,63,69,70} This condition predominantly affects young, active individuals, particularly those engaged in frequent and intensive overhead activities. Adolescents are particularly susceptible due to their rapidly growing and less stable musculoskeletal systems, which may predispose them to hypermobility and instability.

The chronic nature of NTSI and its variability in symptoms often lead to underestimation and misdiagnosis. Unlike acute traumatic cases, NTSI typically presents with more gradual and progressive symptoms, necessitating a thorough clinical evaluation and a multidisciplinary approach. Many patients undergo extensive therapeutic trials and may even face unsuccessful surgeries, ⁴⁷ exacerbating their symptoms and contributing to a cycle of ineffective treatments.

Orthopaedic shoulder tests and diagnostic challenges

Orthopaedic shoulder tests are frequently utilized to diagnose shoulder pathologies, with over 180 tests identified in relevant literature. ⁷⁴ However, the validity and reliability of these tests are often questioned. Many OSTs are based on questionable patho-anatomical theories^74–76 and may overestimate their diagnostic accuracy.^74,77–79 For instance, laxity tests designed to identify joint hyperlaxity are not reliable indicators of instability, as hyperlaxity does not necessarily correlate with symptomatic instability.^56,59,65 Some researchers have begun to use apprehension rather than pain as a criterion for positivity in the sulcus sign, with the idea that it may then hold better diagnostic properties. ¹¹

Moreover, the methodological rigour of many studies is lacking, with small sample sizes and potential biases leading to an overestimation of test efficacy. Meta-analyses and systematic reviews regarding the diagnostic accuracy of OSTs for other joints present similar limitation, reinforcing our confidence about the lack of quality of OSTs for shoulder instability assessments. ⁷⁸ Despite the abundance of tests, only the anterior apprehension test has demonstrated good methodological and reliability metrics for assessing anterior shoulder instability.

In the absence of reliable orthopaedic tests, a thorough subjective evaluation is paramount. Clinicians should also focus on the movements that patients describe as reproducing their symptoms, in an attempt to analyze the motor patterns and potential modifying factors.

Imaging and diagnostic modalities

Imaging plays a limited role as a first-line diagnostic tool for NTSI.^1,72 While it can exclude other pathologies and provide valuable insights into structural abnormalities, imaging findings are often non-specific to NTSI.^63,70,71 The non-routine use of imaging, particularly in the absence of red flags,^59,69–71 has been associated with better patient-reported outcomes and a more accurate understanding of the condition's psychosocial impact. Unnecessary imaging can alter patients’ expectations and beliefs about their pain and their need for surgery, potentially leading to a less conservative and effective management approach, and increased cost.^80,81

If imaging were to be used, radiography would then be used to exclude bony or cartilaginous abnormalities, while CT offers a more detailed bone assessment. Magnetic resonance imaging, especially when combined with arthrography, provides superior visualization of soft-tissue structures and could then provide a detailed assessment of rotator cuff, capsular and labral structures.^59,69

Limitations

Non-traumatic shoulder instability was previously often defined in the literature as MDI (likely due to Matsen's AMBRI classification ⁵ ), whereas MDI is defined by at least two directions if instability. Consequently, we accepted articles with the term MDI but only when it was accompanied by the descriptor ‘non-traumatic’. This variability in definitions may have led us to unintentionally exclude some MDI studies that have included non-traumatic populations. However, we chose to adopt a conservative approach to ensure the quality of our results.

The methodology for the selection of grey literature review is neither exhaustive nor systematic. We adopted and detailed a methodological approach based on recommendations ¹⁸ to ensure reproducibility in article selection; however, the inherent limitations of this method may have introduced some selection bias, which we consider to be minor.

We chose to include literature reviews and theses. The data in these literature syntheses are (at least partially) interpreted by their authors, and there is a risk of double counting if the data presented in these reviews and theses are already included in the original articles we selected. However, our review does not adhere to a purely quantitative paradigm; thus, our results are not weighted by the number of occurrences of a given piece of information. Therefore, the risk of bias related to this methodological aspect is minimal.

Future research directions and methodological improvements

The field of NTSI research faces several challenges, notably the heterogeneity of NTSI forms which complicates classification and study. Historically, NTSI has often been conflated with MDI and other traumatic forms, leading to misclassification and incomplete understanding of non-traumatic instability. To address these issues, future research should focus on:

1. Developing a Unified Definition and Standardized Diagnostic Criteria: Selecting or creating a unique classification system for NTSI to differentiate it from traumatic and multidirectional forms. We plan to launch an international Delphi consensus study to achieve this goal.

2. Improving Methodological Rigour: Strengthening study designs by increasing sample sizes (potentially through multicenter studies to address the low prevalence) to accurately assess the diagnostic properties of orthopaedic tests (e.g., studies evaluating the change in the criterion for positivity from pain to apprehension) and classifications (e.g., the classification emerging from the Delphi consensus), as well as to estimate the true prevalence of NTSI.

3. Evaluating Imaging and Non-Invasive Methods: Investigating how imaging influences the management of NTSI and shapes patients’ perceptions of the need for invasive procedures, while also exploring non-invasive diagnostic methods to minimize unnecessary interventions and costs.