Glenohumeral Osteoarthritis: Disease Burden, Current Understanding, and Gaps in Knowledge and Treatment

Abstract

Background:

Glenohumeral osteoarthritis (GHOA) is a prevalent, costly problem among the aging population. Despite this, the underlying pathobiology and pathomechanics of the disease have not been fully elucidated, and treatment options remain limited to date.

Purpose:

The goal of the 2024 Arthritis Foundation (AF) and AOSSM GHOA Think Tank was to review the present body of knowledge regarding GHOA and identify opportunities for future inquiry to ultimately improve patient outcomes.

Study Design:

Narrative review.

Methods:

This narrative review stems from the evidence-based discussion of the 2024 AF AOSSM Think Tank. This meeting was composed of 4 sessions, on the topics of “Burden of Disease and Shoulder OA,”“Shoulder Conditions and the Development of Shoulder OA,”“Treatment Strategies for Patients With Shoulder OA,” and “Designing and Conducting Multicenter Trials.” Each of these sessions was further divided into 4 lectures pertaining to each topic. Speaker presentations were made available for subsequent review and incorporation of content into this review.

Results:

The epidemiology of GHOA is reviewed, as well as the basic science of arthritis, the roles of repetitive throwing, shoulder instability, and rotator cuff disease in GHOA, along with existing treatment approaches.

Conclusion:

Several unanswered questions are in these fields that warrant future research are highlighted.

The Burden of Disease

Primary osteoarthritis (OA) affecting the glenohumeral joint (GHOA), is a growing problem among the aging population, with a radiographic prevalence of 16% to 20% among middle-aged and elderly adults. ⁴² Shoulder pain is likewise prevalent, with 16% of adults experiencing disabling shoulder pain daily. ⁶⁴ Among patients with atraumatic shoulder pain, primary GHOA is the second most common diagnosis after rotator cuff (RC) disease. ⁹³ Certain demographic risk factors for primary (idiopathic) GHOA include advanced age, female sex, smoking, elevated body mass index (BMI), hypertension, and hyperlipidemia.^93,101 GHOA may also develop secondary to chronic shoulder instability or repetitive overuse in the setting of athletic activities or occupation.^11,56 Finally, untreated RC tears may eventually result in the development of cuff tear arthropathy (CTA), marked by a distinct wear pattern and progression of anterosuperior escape.^29,73,98 The goal of the 2024 Arthritis Foundation (AF) AOSSM Think Tank was to review and synthesize the evidence related to the pathobiology, pathomechanics, and treatment options for GHOA, while identifying gaps in knowledge and areas for future research. The resulting discussion is presented below.

Pathobiology of OA and Targeted Therapies

Joint injury leads to progressive molecular tissue damage and chronic inflammation. The inflammatory process of joint injury leading to chondral breakdown has been well described in the knee OA literature, where this process can be conceptually divided into inflammation of the joint lining, or synovitis, and chondral breakdown.^59-61 Synovitis is associated with elevated expression of interleukin-1 (IL-1) and downregulation of the IL-1 receptor antagonist. ⁵⁸ Chondral breakdown is mediated by numerous proteases and is perpetuated through cellular signaling with inflammatory and immune cells, as well as osteoblasts, osteocytes, and osteoclasts within the subchondral bone. Joint inflammatory markers such as C-terminal cross-linked telopeptides of type II collagen increase after an acute insult to the joint and remain elevated for several weeks after a surgical intervention, before declining to preoperative levels.^41,61 Thus, surgery frequently results in a secondary insult to an inflamed joint. Timing of surgery relative to the acute joint injury is therefore an important clinical consideration. ⁴⁰ Ultimately, these posttraumatic and postsurgical pathways that have been identified in the knee OA literature require future validation in studies of GHOA.

Chronic inflammation and the persistence of inflammatory markers long after injury or surgical intervention have been studied through clinical trials, many of which have focused on knee OA.^31,32 After knee injuries, as many as one-third of patients display signs of chronic inflammation and persistent inflammatory markers. ⁴³ Within the shoulder, there may also be specific inflammatory markers, together marking an “inflammatype,” which have not yet been defined. Future research is necessary to identify biomarkers that may be predictive of subsequent development of GHOA. ¹¹¹

Certain key pathways mediating OA have been identified in the knee OA literature, most notably the canonical Wnt (“Wingless/Integrated”) signaling pathway. Its prolonged activation appears to have a detrimental effect on cartilage. ²¹ Conversely, activation of the TGF-β (transforming growth factor beta) pathway has been shown to play an important role in healthy cartilage homeostasis. ²¹ Inhibition of the Wnt pathway through the small molecular modulator Lorecivivint has shown some early promise in the treatment of knee OA, with patients showing improvement in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain and function subscores at 24 weeks after treatment in a phase 2 randomized controlled trial. ¹⁰⁸ Treatment with the angiopoietin-like 3-derivative LNA043, which inhibits the Wnt pathway by increasing expression of Wnt signaling inhibitors Dickkopf-related protein 1 and frizzled-related protein, was shown to induce expression of hyaline cartilage synthesis genes and inhibit OA progression–related genes over ≥21 days, in a phase 1 randomized controlled trial. ³³ Additional therapeutic targets for improved chondrogenic differentiation include the proteoglycan Agrin, which has been shown to recruit and promote differentiation of chondrogenic growth/differentiation factor 5 (GDF5)-lineage joint-resident progenitor cells through inhibition of the Wnt pathway (Table 1).

Table 1.

Summary of Biomolecular Factors Involved in Osteoarthritis

Name	Abbreviation	Role in Osteoarthritis
Interleukin-1	IL-1	Cytokine elevated in the setting of synovitis; marker of inflammation associated with osteoarthritis
Interleukin-1 receptor antagonist	IL-1RA	Cytokine that binds to the IL-1 receptor and prevents IL-1 from binding, preventing signaling through this pathway
C-terminal cross-linked telopeptides of type II collagen	CTX-II	Biomarker that is associated with cartilage breakdown
Wingless/Integrated	Wnt	Signaling pathway involved in cartilage homeostasis; prolonged activation may have a detrimental effect on cartilage
Transforming growth factor beta	TGF-β	Signaling pathway involved in cartilage homeostasis; activation promotes cartilage maintenance
Dickkopf-related protein 1	DKK-1	Inhibitor of canonical Wnt signaling
Frizzled-related protein	FRB	Inhibitor of canonical Wnt signaling
Runt-related transcription factor	Runx	Family of transcription factors downstream of TGF-β signaling; Runx2 is associated with hyaline cartilage production
Core binding factor beta	CBFβ	Subunit of a core-binding transcription factor involved in chondrocyte and osteoblast differentiation

Runx (Runt-related) transcription factors are downstream targets of TGF-β signaling. ³⁰ Activation of Runx2 transcription to increase hyaline cartilage production is another therapeutic target under early investigation in the setting of OA treatment. ⁶³ Sprifermin is a recombinant human fibroblast growth factor–18 analog, which can bind to the fibroblast growth factor receptor 3 receptor and in turn activate Runx2 transcription.^39,63 Preclinical trials of Sprifermin have demonstrated an increase in hyaline cartilage and extracellular matrix production with treatment. In a subgroup analysis of the FGF-18 Osteoarthritis Randomized Trial with Administration of Repeated Doses knee OA trial, Sprifermin treatment resulted in a dose-dependent improvement in WOMAC pain scores as well as increased cartilage thickness on magnetic resonance imaging (MRI) at 3 years after initiation of treatment. ³⁹ Further studies have demonstrated that the Runx partner core binding factor β (CBFβ) is pivotal in chondrocyte and osteoblast differentiation and is decreased in expression in aged mouse articular cartilage and in human OA cartilage.^19,20,110 Kartogenenin is a small molecule that regulates the CBFβ-Runt-related transcription factor 1 (Runx1) transcriptional program and enhances the synthesis of cartilage matrix components including type II collagen and aggrecan. ⁴⁸ Further clinical trials are required to investigate these signaling pathways and to establish the long-term efficacy of these molecular approaches to treating glenohumeral joint OA.

Scapular Morphology and GHOA

Dr Gilles Walch described 3 major variants of primary GHOA based on glenoid morphology: type A, marked by a well-centered humeral head with concentric erosion of the glenoid; type B, marked by posterior subluxation of the humeral head corresponding to eccentric wear of the glenoid posteriorly; and type C, marked by glenoid retroversion of >25°. ⁹⁹ Correlation of glenoid morphology with RC muscle atrophy measured on MRI demonstrated that B-type glenoids, excessive retroversion, and posterior humeral head subluxation are associated with imbalances in RC muscle volume.^18,38 An increased infraspinatus to subscapularis ratio was likewise directly correlated with a B2 glenoid (defined by a biconcave, retroverted morphology), increased retroversion, and increased humeral head subluxation in a multicenter cohort. ¹ In addition to cuff muscle asymmetries, increasing posterior deltoid muscle area has been correlated with increased retroversion in B2 glenoids. ⁷⁷ In patients who have glenohumeral instability, the subscapularis to infraspinatus/teres minor muscle area ratio has also been correlated with the direction of shoulder dislocation. ⁴⁴

Whether these muscular imbalances are the primary cause of the pattern of GHOA wear or resultant effects of preexisting bony anomalies remains unknown. ¹⁰⁶ A recent study found that B2 glenoids display significantly greater premorbid retroversion, as defined by referencing the anterior glenoid/paleoglenoid, compared with nonarthritic shoulders. ⁵⁵ Acromial dysplasia may represent an additional contributing factor to arthritic wear patterns, as posterior rotation of the coracoacromial arch is associated with shoulder instability.^17,45 Thus, we conclude that scapular morphology and the surrounding shoulder musculature may have a multifactorial effect on eccentric wear patterns in GHOA, with muscle imbalances, premorbid retroversion, and acromial dysplasia all potentially playing a role.

GHOA in Throwing Athletes

Repetitive high-intensity overhead activities correlate with specific degenerative changes observed in the shoulders of elite athletes. ⁸⁵ Javelin throwers have been shown to have high rates of internal rotation loss, partial-thickness or full-thickness RC tears, superior migration of the humeral head, and degenerative changes within the glenohumeral joint that correlate in severity with the duration of the high-performance phase of their careers. ⁹⁰ Former collegiate and professional baseball players have demonstrated persistent shoulder joint pain in 28% of 117 athletes surveyed. ¹⁴ The phases of pitching most associated with injury to the shoulder are the late cocking and deceleration phases. ²⁶ In the late cocking phase, internal impingement may occur between the posterosuperior glenoid labrum and the articular side of the posterosuperior RC and greater tuberosity. ¹⁰⁰ This process eventually leads to labral pathology and articular-sided RC tears. In certain cases, extensive peel-back of the labrum may result in a type VIII superior labrum from anterior to posterior tear. ¹⁵ The deceleration phase of throwing places the greatest forces on the RC and capsule as the muscles contract eccentrically to reduce the velocity of the arm. ²

Repetitive throwing has also been associated with insufficiency of the anterior capsuloligamentous structures of the glenohumeral joint. ³⁵ Previous studies have noted significant anterior translational forces sustained during the late cocking/early acceleration phases of throwing, in addition to an increase of ~18° in external rotation corresponding to a 30% increase in anterior capsular laxity due primarily to elongation of the anterior inferior glenohumeral ligament (IGHL).^68,91 Glenohumeral internal rotation deficit and scapular dyskinesia may result from poor throwing mechanics and a lack of stretching, in addition to poor core strength. ³⁵

In summary, the literature associating overhead throwing with GHOA remains limited; however, the available evidence suggests a concerning link to early degenerative changes that may progress to GHOA in the long term.

GHOA in the Unstable Shoulder

Acute bony and cartilaginous lesions resulting from glenohumeral dislocations may ultimately progress to GHOA, although the natural history of this process remains unclear. Habermeyer et al ³⁶ described a stepwise progression of labroligamentous injury with an increasing number of instability events. This injury cascade begins with a Bankart lesion, followed by an IGHL attachment rupture (double lesion/Perthes), which may then progress to an anterior labral periosteal sleeve avulsion (triple lesion [ALPSA]). The process culminates in both labral and capsular degeneration (quadruple lesion). ³⁶ Compared with isolated Bankart lesions, ALPSA lesions are associated with 3 times the number of instability events before surgical intervention and carry a higher surgical failure rate of 17.4% to 19.2% versus 7.4% for isolated Bankart tears. ⁷⁹ ALPSA lesions are also more likely to be associated with other intra-articular pathology, including cartilage lesions of the joint. ³

Glenolabral articular disruption (GLAD) lesions are acute cartilage injuries of the glenoid resulting from traumatic instability, which if left untreated may place patients at high risk for progression to arthritis. ⁸¹ GLAD lesions carry a significantly higher risk of failure of arthroscopic treatment than Bankart lesions alone, which may in part be due to their biomechanical effect on translation of the humeral head. ¹⁰⁵ The incidence of acute glenoid chondral lesions in anterior shoulder instability patients ranges from 36% to 57%, compared with a 72% incidence of an anterior labral tear, and a 46% incidence of Hill Sachs lesions.^75,87 The Multicenter Orthopaedic Outcomes Network shoulder instability cohort revealed overall rates of bone or cartilage lesions of 18.4% in patients undergoing primary anterior stabilization, and 47.6% in patients undergoing revision surgery. ²⁷ Risk factors independently associated with the presence of bone and cartilage lesions included male sex, revision surgery, Black race, higher BMI, increasing patient age, and lower preoperative 36-Item Short Form Health Survey Physical Component Summary. ²⁷ Additional evidence has shown that the risk of failure of arthroscopic stabilization increases with time from first dislocation >6 months and male sex. ⁸²

Treatment options for acute bone and cartilage injuries sustained in the setting of glenohumeral instability range from nonoperative measures, including nonsteroidal anti-inflammatory drugs, physical therapy (PT), and injections, to operative treatment. Surgical treatment may include debridement of the lesion, advancement of the labrum over the lesion, microfracture, osteochondral autograft transfer system (OATS), osteochondral allograft transplantation (OCA), or autologous chondrocyte implantation (ACI). Microfracture has shown a long-term success rate of 67% to 75%.^89,102 However, in a study of military patients who underwent microfracture and labral repair compared with patients without an osteochondral injury who underwent labral repair alone, there was a lower return to active duty of 54% in the microfracture group versus labral repair alone (94%). Additionally, the microfracture group had worse patient-reported outcomes (PROs), suggestive of the negative effect of an osteochondral injury in this population. ³⁴

A case series of 7 patients who underwent OATS for full-thickness cartilage lesions of the shoulder demonstrated improved Constant scores at a mean of 8.75 years after surgery, with healing and a congruent joint surface in all but 1 patient. ⁵³ A case series of 20 patients who underwent OCA to the humeral head demonstrated incorporation in 18 out of 20 grafts and a high satisfaction rate (9 of 9 patients) among those who underwent an isolated humeral OCA, with a low satisfaction rate in patients with bipolar lesions (2 of 11 patients satisfied). ⁸⁶ A case series of 4 patients who underwent ACI for humeral cartilage defects demonstrated an improvement in PROs at a mean follow-up of 41 months, with follow-up MRI demonstrating satisfactory defect coverage. ¹³

In a study assessing the long-term development of GHOA after open stabilization surgery with 5- to 20- year follow-up, the onset of OA correlated with the number of instability events preoperatively. ⁷⁶ In a second study of the presence of glenohumeral osteoarthrosis in 570 shoulders after stabilization surgery, the overall rate of postoperative OA over the 6.5-year mean follow-up period was 19.7%. OA correlated with older age at the time of surgery, increased number of preoperative dislocations, and longer follow-up. ¹⁶ Early intervention with arthroscopic stabilization in patients who sustained a first-time shoulder dislocation before age 25 has been shown to reduce the risk of secondary dislocation and to improve functional outcomes compared with nonoperative treatment over a 2-year follow-up period. ⁸³

Compared with anterior instability, outcomes after treatment of posterior instability with respect to the development of GHOA remain less certain. Multiple studies have found that while surgical intervention for posterior instability is effective at treating pain, it has also been correlated with progression of GHOA compared with nonoperative care.^62,96 Posterior shoulder dislocation events carry a high risk of posterior bone loss development, with each instability event being responsible for up to 5% bone loss. ⁷ Moderate posterior glenoid bone loss, defined as >13.5%, is associated with an 88% presence of pain at final follow-up and a 33% revision surgery rate. ¹⁰⁷ Surgical management of posterior instability is largely dependent on bone loss, with glenoid bony augmentation recommended in the primary setting with >20% bone loss, or in the revision setting with 10% to 20% bone loss. ²⁵

In conclusion, a growing body of evidence suggests that early surgical intervention for shoulder instability is likely warranted, as it may prevent subsequent dislocations, bone loss, and ultimately, GHOA. Although the connection between an isolated traumatic shoulder dislocation and GHOA remains unclear, recurrent dislocations resulting in progressive chondral injury and bone loss accelerate the process of posttraumatic GHOA.

RC Tears and GHOA

Acromial morphology may play a role in determining which patients are at risk for RC tendon tears and which are at risk of developing primary GHOA. The critical shoulder angle, measured between the glenoid fossa and the lateral border of the acromion on a Grashey radiograph, is elevated in patients with degenerative RC tears and reduced in patients with primary GHOA. ⁹⁴ The acromial index is also elevated in the setting of cuff tears compared with GHOA, while the lateral acromial angle has been found to be reduced. ¹⁰⁹ A study exploring 3-dimensional modeling of scapular morphology determined that patients with cuff tears exhibited higher glenohumeral joint forces and a higher instability ratio in abduction, while patients with GHOA had a higher shear-to-compressive force ratio compared with RC tear patients. ⁷⁸ Taken together, these data suggest that acromial and scapular morphology may dictate distinct degenerative pathologies of the shoulder and that the predisposing anatomic factors for RC tears and primary GHOA are separate, despite the shared demographic features among patients with these conditions.

Both increased bulk of the posterosuperior cuff muscles and fatty infiltration of those same muscles have been independently correlated with increased retroversion of the glenoid and humeral head subluxation. ⁷² This suggests that the RC may potentially have both a primary role and a compensatory response to the development of glenoid deformity.

Although the role of RC pathology in primary GHOA may be ambiguous, the effect of untreated, large-to-massive RC tears in the development of secondary GHOA, or CTA, is well-described.^29,73,98 The superior migration of the humeral head that may occur with a massive RC tear was described as early as 1793, and the term “cuff tear arthropathy” was coined by Charles Neer. ⁷³ The pathogenesis of arthritis in CTA is thought to be a combination of mechanical factors and nutritional factors that ultimately lead to destruction of the chondral surfaces after anterosuperior escape of the humeral head, in addition to remodeling of the acromion and proximal humerus. ⁸⁸

Relative to the high prevalence of degenerative RC tears, the rate of CTA developing from these tears is much lower, estimated by Neer to occur in roughly 4% of all cuff tears. ⁸⁴ Similarly, not all patients with massive RC tears will progress to CTA. While it seems that patients with massive cuff tears who maintain their range of motion >90° in all planes may be at less risk for progression to CTA than those with pseudoparalysis, advanced muscle degeneration, or those who have failed an RC repair, it remains unknown which patients will progress to CTA based on the diagnosis of a massive RC tear alone.^49,70

The clinical outcomes of primary GHOA on the treatment of RC tears has been studied, as has the effect of RC tears on shoulder arthroplasty. In patients with mild GHOA, defined as Samilson and Prieto grade 1, who underwent repair of large-to-massive RC tears, there was no change in failure rate, pain, or functional outcomes at 2 years after surgery. ⁴⁶ In a study utilizing the Shoulder Osteoarthritis Severity score to quantify global shoulder pathology as it relates to postoperative changes in American Shoulder and Elbow Surgeons (ASES) scores after RC repair, the authors found that worse overall intra-articular pathology correlated with decreased improvement in ASES scores after surgery. ²² When assessing the effect of RC tears on patients undergoing reverse total shoulder arthroplasty (RTSA), patients who had primary GHOA with an intact RC had higher Constant scores at 2 years after surgery than patients with primary GHOA and cuff lesions, or patients with CTA who were undergoing RTSA. ⁷⁴ Taken together, these data suggest that outcomes after a joint preservation surgery such as RC repair are negatively affected by the presence of early GHOA, and likewise, outcomes after RTSA can be negatively affected by the presence of CTA, even though this remains among the most common indications for RTSA.

The optimization of RC muscle quality via the prevention of atrophy and fatty infiltration is a growing area of interest within the field, as poor muscle quality is linked to failed repairs, which may in turn progress to CTA. ⁶ Certain pharmacologic agents, including the FDA-approved β3-receptor agonist mirabegron, have shown promise in small animal studies of RC tears by halting the progression of muscle fatty infiltration and fibrosis via beige and brown adipose differentiation of mesenchymal progenitor cells within muscle. ¹⁰³ Blood flow restriction (BFR), in combination with low-intensity exercise, has also been investigated as a therapeutic approach to increasing muscle strength around the shoulder, with mixed results.^12,50,57 High-quality clinical trials of pharmacologic agents such as mirabegron and rehabilitation techniques such as BFR are needed to better assess their potential efficacy in the treatment of RC muscle degeneration and the downstream effects on GHOA.

GHOA Treatment Options

Physical Therapy

The physical impairments caused by GHOA can be divided into 4 treatment categories for PT: muscle strength, range of motion, movement impairment (as a factor of pain), and night pain (sleep disturbance). When treating a patient with PT for GHOA, it is important to modify activity based on the pain-monitoring model, ensuring that pain levels after PT return to preactivity levels by the morning after a therapy session. ⁹⁵ In patients with excessive pain in the setting of GHOA, modalities such as ice, heat, and high-intensity sensory transcutaneous electrical nerve stimulation can be applied for pain control as an adjunct to therapy. When performing stretching exercises for GHOA, it is important to respect bony deformity that is limiting range of motion, as attempting to push past these structural limitations will result in excessive pain and possible injury. Encouraging continued therapeutic exercise in the setting of painful OA is recommended, as it has been shown to aid in pain relief and is safe for articular cartilage. ¹⁰ In spite of these recommendations, the evidence for PT in the treatment of primary GHOA remains insufficient, largely owing to a lack of high-quality clinical trials, although some remain ongoing. ⁵² The effect of PT on GHOA therefore represents an understudied area that warrants further longitudinal observational studies to better identify which patients may best respond to nonoperative care.

Factors beyond the disease burden of GHOA may additionally play a role in patient outcomes, including mental health, resilience, and level of preoperative education provided to patients.^23,66,97 In particular, patients with worse mental health status are more likely to have poor outcomes after shoulder arthroplasty. ⁴⁷ These findings highlight the need to provide feasible interventions to address mental health and resilience and to ensure adequate preoperative education in patients who are being treated for GHOA, whether with PT or surgery.

Injections

Corticosteroid injections are a mainstay of nonoperative care for GHOA, although limited data exist regarding their efficacy. A prospective cohort study of 29 patients (30 shoulders) who underwent corticosteroid injections for GHOA and were followed over a 1-year period found that patients with more severe preinjection shoulder dysfunction observed greater symptomatic relief than patients with milder dysfunction. Improvements in pain and function were limited in duration to 4 months after injection. ⁶⁷ In the setting of shoulder arthroplasty, several recent studies have found an association between corticosteroid injections and increased risk of prosthetic joint infection, with risks of infection up to 9.0% in patients who underwent injections within 1 month of surgery.^5,92

A systematic review of 1879 patients treated nonoperatively with a median 6-month follow-up reviewed the efficacy of other injectable therapies, including hyaluronic acid (HA), bone marrow aspirate concentrate (BMAC), leukocyte-rich and leukocyte-poor platelet rich plasma and autologous conditioned serum. ⁵¹ Thirteen of the 19 studies analyzed studied HA injections, and the majority of studies reported a short-term improvement in visual analog scale scores, although one study found no difference between HA injection and placebo saline control. ⁵¹

In a double-blind randomized controlled trial of leukocyte-poor platelet-rich plasma compared with HA injection for GHOA, there were no significant differences in PROs or functional scores between groups over the 12-month follow-up period. Both groups demonstrated some improvement in pain scores beginning at 1 to 2 months after injection. ⁵⁴ A pilot randomized controlled trial of 25 shoulders treated with BMAC injection versus cortisone for GHOA found that BMAC resulted in improved PROs compared with cortisone over the 12-month follow-up period, with the exception of the Western Ontario Osteoarthritis of the Shoulder index, which did not differ between groups. ²⁸ Overall, intra-articular injectable therapies represent viable short-term treatment options for symptoms of GHOA; however, further high-quality evidence is needed to evaluate their true efficacy.

Arthroscopy

Arthroscopic debridement of the shoulder for GHOA is a treatment option that has been investigated primarily for younger patients with early GHOA, or those who otherwise are not candidates for shoulder arthroplasty but have failed nonoperative management. In a case series of 25 patients with a mean follow-up of 34 months who underwent glenohumeral joint debridement, loose body removal, partial synovectomy, and subacromial bursectomy, 8% had results that were deemed excellent, 72% had good results, and 20% had unsatisfactory results, with 83% of patients who preoperatively had severe stiffness demonstrating an improvement in motion postoperatively. ¹⁰⁴ More recently, Millett et al ⁶⁹ described the approach of Comprehensive Arthroscopic Management (CAM) as a way of arthroscopically addressing known pain generators in the arthritic shoulder. This approach involves debridement of labral tears and loose chondral flaps, loose body removal, and a complete capsular release, humeral head osteoplasty, and axillary nerve neurolysis.^4,69 CAM has demonstrated a survivorship of 75.3% at 5 years and 63.2% at ≥10 years postoperatively, with humeral head flattening and severe joint incongruity identified as risk factors for failure. ⁴

Arthroplasty

Shoulder replacement remains the standard of care for painful end-stage GHOA that has failed nonoperative treatment. Over the past decade, shoulder arthroplasty rates have dramatically increased,^8,24 with a 3-fold increase observed in RTSA, a 1.5-fold increase in anatomic total shoulder arthroplasty (TSA), and a decline in hemiarthroplasty. ³⁷

Large retrospective database studies have identified an increase in perioperative adverse events associated with RTSA compared with TSA that cannot be explained by patient demographics or comorbidities. ⁹ Overall, surgeon volume >10 shoulder arthroplasties per year has been correlated with lower odds of revision surgery within a 2-year follow-up period. ⁹ In the United States, approximately 75% of all shoulder arthroplasties are performed by surgeons who perform <10 TSAs or RTSAs per year. ⁹ Certain strategies have been developed to control for surgeon experience with shoulder arthroplasty, including computed tomography–based planning, patient-specific guides, augmented reality assistance, and robot-assisted total shoulder systems. Observation of the effect of scapulothoracic orientation on postoperative patient outcomes may likewise allow for further optimization of component positioning based on patient posture and anatomy. ⁷¹ Future studies are needed to assess both the short-term and the long-term efficacy of these recent strategies to standardize outcomes for shoulder arthroplasty.

The postoperative complication profiles of both TSA and RTSA have evolved with advancements in technology, particularly related to implant design and materials science. For TSA, the leading complications are RC failure, glenoid component loosening, and infection. ⁸⁰ For RTSA, the most common complications include acromial/scapular fracture, instability, and persistent pain. ⁸⁰ However, it should be noted that overall complication rates for each procedure have decreased substantially over the past 10 to 15 years. ⁶⁵

Inspiration for Future Multicenter Studies

Based on the review and discussion of the evidence above at the 2024 AF AOSSM GHOA Think Tank, the following questions have been posited as critical knowledge gaps that warrant future multicenter cohort studies and clinical trials:

1. What is the role of novel small-molecule therapies and recently developed pharmacological agents in the treatment of GHOA?

2. What is the natural history of traumatic shoulder instability as it relates to the pathogenesis of GHOA?

3. What are the defining risk factors for patients with RC tears that lead predictably to secondary GHOA?

4. What is the efficacy of PT in the treatment of GHOA?

5. What factors account for the increase in higher perioperative morbidity or better clinical outcome with RTSA versus TSA?

In formulating these unsolved queries, we hope to spark future multicenter collaborations that may further our understanding of GHOA to improve treatment options and patient outcomes.