The presence of radiographic glenohumeral osteoarthritis does not necessarily translate into clinically significant osteoarthritis [17]. Hovelius et al. in their series of 255 patients followed during 25 years reported that no patient had surgery because of arthropathy or any other disorder (except for revision due to recurrent instability);[4]. Of them, only seven patients presented arthropathy that was somehow disabling, mainly because of pain. Most patients often reduce their sporting activity or change their working routine in order to minimize the effect of the symptoms.

In general, symptoms might not differ from those of patients suffering primary osteoarthritis of other origin. They typically complain of chronic pain with an insidious onset, predominantly in the morning and exacerbated with weather changes. Regarding range of motion, stiffness is frequent. There can be a loss of active and passive external rotation at 90° of abduction and also in neutral position. Whether this is a cause of excessive tightening of anterior structures in previous surgeries or is due to the articular changes that occurred in degenerative disease is still controversial [2]. Audible “clicks” with shoulder motion may indicate bursitis, biceps tendon pathology, osteophytes, or loose bodies. Ellman described a “compression-rotation” test for the arthritic shoulder [18]. The patient is placed in the lateral decubitus position with the affected side up. Then, the humeral head is compressed into the glenoid, and the shoulder is internally and externally rotated. Pain is elicited as the arthritic glenohumeral joint surfaces are compressed together. A subacromial injection can be performed first in order to eliminate subacromial pathology as a source of pain and increase sensibility.

Supraspinatus and infraspinatus fossae should be inspected to look for any rotator cuff atrophies. Shoulder altered kinematics or the presence of a scapular dyskinesia should also be ruled out. A routine physical examination for rotator cuff pathology, acromioclavicular joint osteoarthritis, and shoulder instability is usually recommended. In patients over 40 years old with an instability episode, posterior structures are more likely to fail, secondary to pre-existing rotator cuff weakness [19]. In patients over 60 years old, recurrent instability after an anterior dislocation might be caused by a failure of the posterior rotator cuff [20]. The overall frequency of rotator cuff tears after an anterior dislocation ranges between 7 and 32% and rises with advancing age. In these cases, the patient will present with a limited active forward elevation and abduction. Differential diagnosis should include axillary nerve damage.

Treatment for dislocation arthropathy depends on the presentation and the disability that it causes to the patient. Initial nonoperative measures include anti-inflammatory medications, moderate exercise, physical therapy, and injections. In the cases where a previous surgery was performed, it is mandatory to check if metal hardware is responsible for the symptoms, and sometimes, removing it can avoid progressing symptoms.

Speigl et al. recently published their results of different surgical interventions in the context of glenohumeral osteoarthritis. In this study they compared an arthroscopic approach versus a total shoulder arthroplasty (TSA), and the principal outcome measure was total remaining quality-adjusted life years after each treatment option [21]. They concluded that arthroscopic management was the preferred strategy for patients younger than 47 years, TSA was the preferred strategy for patients older than 66 years, and both treatment options were reasonable for patients aged between 47 and 66.

Arthroscopic management of osteoarthritis would include debridement, removal of loose bodies, synovectomy, osteoplasty, and contracture release [22, 23]. The Comprehensive Arthroscopic Management (CAM) procedure was described by Millett et al. [3]. First, a glenohumeral debridement of degenerative labral tissue and unstable cartilage fragments is performed, and loose bodies are excised. A limited synovectomy can also be performed with the use of the shaver or with the radiofrequency probe, and stable chondral lesions can be treated with microfracture. In the cases where there is a significant inferior osteophyte, it can be excised with a high-speed burr and arthroscopic shaver using a posterosuperior portal for visualization. Internal and external rotation of the arm can help in identifying the spur, and fluoroscopy can be used to ensure adequate bone resection. The capsule release is performed at the end of the procedure as it keeps the axillary nerve out of danger during osteophyte excision and improves visualization of the axillary poach. It can be performed with arthroscopic scissors and a monopolar radiofrequency probe starting from inferior and then complete the anterior and posterior release in a standard fashion. Neurolysis of the axillary nerve can be also performed in patients with posterior or lateral pain, compressive signs detected in the MRI, or direct encroaching seen in the inferior poach during arthroscopy. It is performed from proximal to distal to avoid damage to the nerve branches. Other surgical gestures depending on patient’s physical examination can include subacromial decompression, biceps tenotomy or tenodesis, or resection of the acromioclavicular joint.

This procedure provides pain relief, an increase in range of motion, and better functional scores. However, in Millett’s series 6 out of 29 patients were dissatisfied with the result and finally underwent TSA at a mean of 1.9 years after the arthroscopic debridement. This group of patients accounted for a lower preoperative ASES score and had less preoperative joint space. A joint space of less than 2 mm increased 7.8 times the risk to progress to an arthroplasty. They also observed that patients with chondral damage grade IV had worse results and those with a more restricted preoperative external rotation were more satisfied with final outcome. The mean survivorship was 95.6% at 1 year, 86.7% at 3 years, and 76.9% at 5 years [24]. Other authors, however, have published worse results with a return to preoperative pain and range of motion levels within 3.8 months after the surgery [22]. Skelley et al. published a rate of unsatisfactory results in 60.6% of their patients, with a TSA conversion rate of 42.4% at an average of 8.8 months after arthroscopy [22].

For isolated unipolar lesions, repair of the chondral defect management with microfractures, cartilage transplantation, osteochondral allografts, or partial resurfacing has been proposed, but to date the results of these procedures are very scant [25, 26] (Fig. 30.2).

With regard to prosthetic options, it is well known that TSA provides pain relief and improves shoulder function [27]. However, there is a risk of component wear and loosening and also a recommendation for activity restriction that is usually a concern for active patients. As mentioned before, the symptoms of shoulder arthritis in the context of shoulder instability are mild. This fact often leads the patient to a joint-preserving treatment option even in patients aged over 50 years [4, 28]. Although hemiarthroplasty (HA) could be an intermediate solution, it has been demonstrated that it provides less pain relief and functional improvement than TSA. In younger patients, hemiarthroplasty with biologic glenoid resurfacing as an interposition arthroplasty has been performed using different tissues: fascia lata autograft [29], anterior capsule, lateral meniscus allograft [30], and Achilles tendon allograft [31]. In order to avoid concerns about durability of soft tissue interposition, a concentric glenoid reaming (ream and run) was also proposed [32]. The results of these procedures have been inconsistent (Fig. 30.3).

The results of TSA in patients with dislocation arthropathy are reproducible and satisfactory. Matsoukis et al. reported on the results of TSA in 55 patients with a prior shoulder dislocation [33]. They observed improvements in Constant Score and range of motion, and most of the patients rated the result of their surgery as good or excellent. They did not find differences in between patients with and without previous surgeries for their instability. Negative prognostic factors included older age at the time of the initial dislocation and the presence of a rotator cuff tear. Green and Norris also proved an increase in shoulder function and pain relief [34]. There was a revision rate of 3 out of 19. The authors found a more severe arthritis in patients that had previously undergone a nonanatomic procedure and were characterized by a severe internal rotation contracture and subsequent posterior glenoid wear, making the surgery challenging. Sperling et al. reported on 31 cases with a mean age of 46 years consisting of 21 TSA and 10 HA. There was a significant pain relief and increase in abduction and external rotation. However, eight of the TSA and three of the HA required revision surgery [27]. Thus, it is important to mention that results of TSA in this pathology have been proven to be less satisfactory than in primary osteoarthritis [33, 34]. Contracture of the anterior soft tissues and erosion of the posterior glenoid can be related to an increased risk of revision arthroplasty [17, 27].