Degenerative osteoarthritis (OA) of the glenohumeral joint may be a primary, idiopathic process. However, it can also be secondary to a number of inciting factors including post-traumatic, previous capsulorrhaphy, chondrolysis from intra-articular pain pumps, hardware complication, persistent instability, and developmental problems that accelerate the degenerative process.
Although both OA and inflammatory arthropathies such as rheumatoid arthritis lead to a final common pathway of cartilage loss and joint destruction, these should be noted as distinct entities with different clinical profiles and treatment options.
Patients classically present with pain and stiffness. Detailed history typically reveals pain that is aggravated with movement and/or weight-bearing. Although the onset of pain can be associated with a traumatic incident, most will report progressive and deteriorating pain without trauma. Pain can be present at night-time, and in cases of end-stage OA, cause disturbances in sleep. Associated symptoms may include joint swelling, auditory crepitus, and paresthesias secondary to compression from osteophytes. Due to these symptoms, patients often report limitations in their shoulder range of motion and ability to perform routine activities of daily living with the affected arm.
Examination must first address possible radicular pain to the shoulder that actually originates from the cervical spine. Therefore, a detailed cervical spine examination as well as neurological examination of the involved upper extremity must be completed and documented.
Findings from visual inspection of the shoulder may include joint swelling and disuse atrophy. Tenderness may be elicited over the anterior or posterior joint line. In addition, tenderness may be noted about the acromioclavicular joint and the bicipital groove suggesting associated pathology about these structures.
Both passive and active range of motion may be limited, particularly external rotation. Active motion is also often associated with pain and crepitus about the joint. In neutral position, testing of the rotator cuff should demonstrate reasonable strength. However, with advanced disease, these maneuvers may also elicit pain which will result in “giving way” and the appearance of weakness.
Radiographs represent the primary means to establish the diagnosis of glenohumeral OA. Characteristic finding include joint space narrowing with subchondral cysts and sclerosis. Minimum views required for assessment of the joint are the anterior posterior (AP) and axillary views. Other views, such as the “Y” lateral or Grashey view may also be helpful in characterizing the joint. In more advanced OA, radiographs will reveal inferior humeral head osteophytes (“bearded goat”) and posterior glenoid wear with posterior humeral head subluxation (Figure 1).
If radiographs do not provide sufficient information, computed tomography (CT) may be used. CT scans, especially with 3-D reconstructions, can provide accurate assessment of the bony anatomy. Therefore, CT scans are generally obtained in cases with significant bony changes (e.g., glenoid erosion) and are routinely utilized by some surgeons for pre-operative planning purposes.
Magnetic resonance imaging (MRI) is not necessary for evaluation of patients with glenohumeral OA. However, it can be quite valuable in patients with suspected concomitant pathologies about the soft tissues, especially the rotator cuff. In additional, degenerative tears of the labrum and possibly the biceps tendon can also be confirmed to identify other potential sources of pain about the joint.
The mainstay of non-operative management for glenohumeral OA is pharmacotherapy with acetaminophen and non-steroidal anti-inflammatory drugs to minimize the pain associated with the disease. Physiotherapy can be helpful in maintaining the strength and motion about the joint. However, excessive therapy exercises should be avoided as movement of incongruent surfaces can result in exacerbation of symptoms.
In patients with severe pain, corticosteroid injections can be considered. There are, however, reports of collagen tissue damage following steroid injections. Therefore, these injections must be utilized judiciously in order to avoid further damage to the joint, especially to the rotator cuff tendons. In addition, reports have also suggested that “in-office” injections without guidance demonstrate high rates of missing the glenohumeral joint.
Recently, viscosupplementation with injections of hyaluronic acid has been proposed as another non-operative treatment option for glenohumeral OA. To date, however, there is only sparse data on the efficacy of such treatment. Therefore, its use has not been approved by the FDA for this indication.
Indications for Surgery
Pain refractory to non-operative care is the main indication for surgical management. Additionally, surgery is indicated when the disease has a significant impact on the patient’s activities of daily living and quality of life.
Surgical options include:
Arthroscopic debridement of shoulder
Hemiarthroplasty with possible glenoid “reaming” as needed.
Total shoulder arthroplasty
For patients with mild to moderate degenerative changes, arthroscopic debridement of the glenohumeral joint may provide symptomatic improvements. It should be noted, however, that debridement is unlikely to alter the disease process and may only be a temporizing measure.
Arthroscopy can be set up in the beach chair or lateral decubitus positions. Primary goals of the arthroscopy are to re-balance the joint and remove the osteophytes. Typically, the anterior capsule is contracted and has to be released. This must be performed while carefully protecting the subscapularis tendon. In addition, the concavity of the glenoid should be restored. In patients whose glenoid is not concave, it may need to be “re-shaped” to accommodate the humeral head. The inferior humeral head osteophytes are difficult to reach with standard arthroscopy portals and will often require a secondary portal in the posterior inferior quadrant of the joint. It should be noted that the axillary nerve is located just inferior to the joint capsule and must be carefully protected during this portion of the procedure. Prior to completing the arthroscopy, any other sources of pain should be identified and addressed. These may include debridement of degenerated labrum, repair of the rotator cuff, biceps tendon tenotomy or tenodesis, acromioplasty, or distal clavicle excision.
For young patients with end-stage OA, arthrodesis of the shoulder joint remains a viable option. Although functional capacity is limited, this procedure will typically result in pain free use of the arm. The position for the fusion has been recommended, somewhat arbitrarily, to be in 30 degrees of flexion, 30 degrees of internal rotation, and 30 degrees of abduction. It is important to maximize the potential use of the arm. Therefore, the arm position must be checked intra-operatively to ensure that the hand can easily reach the mouth and the anterior perineal area for feeding and hygiene purposes, respectively. This may require the arm be fused in slightly more internally rotated, but less abducted, position. There are also various patterns of hardware construct to optimize fusion. After contouring the humeral head to match against both the glenoid and the undersurface of the acromion, the hardware should provide rigid fixation while compressing the bony surfaces.
Another option for young patients with end-stage OA is shoulder resurfacing (Figure 2). This is typically performed through a deltopectoral approach and the humeral head is resurfaced with a metal cap. On the glenoid side, if the remaining cartilage is viable, no treatment may be needed. With advanced degeneration of the glenoid, biological resurfacing with various graft options including anterior capsule autograft, Achilles tendon allograft, lateral meniscus allograft have all been described. Advantages of this procedure include retention of bone stock for possible future revision surgeries. Primary disadvantages of this procedure include unpredictable outcome. In addition, recent literature also suggests that, even after good initial outcome, the result can deteriorate over the next 4-5 years.
For older patients with end-stage osteoarthritis of the glenohumeral joint, shoulder arthroplasty is the treatment of choice (Figure 3).
Options for shoulder arthroplasty include humeral head replacement alone (hemiarthroplasty – HA), humeral head replacement with glenoid reaming, or humeral head replacement with glenoid prosthestic resurfacing (total shoulder arthroplasty – TSA).
Primary goals of shoulder arthroplasty are to recreate the native anatomy of the shoulder joint. Therefore, normal retroversion of the humeral head, relative height of the humeral head with respect to the greater tuberosity, neutral version (relative to the scapula) of the glenoid surface, and centering of the humeral head on glenoid surface must all be restored. In addition to recreating the bony anatomy, the shoulder joint also must be balanced which typically requires anterior and inferior capsular release, and possibly, posterior capsular imbrication.
In comparison to HA, multiple studies have reported that patients with TSA demonstrate more predictable pain relief with improved motion. Therefore, TSA is generally favored over HA for most patients with endstage glenohumeral OA. In patients whose glenoid cannot tolerate a prosthesis (due to small size or excessive bone loss) or in patients with an irreparable rotator cuff tear that cannot be treated with a reverse total shoulder arthroplasty, HA may still be indicated. Some recent studies suggest that HA with glenoid reaming to induce a concentric surface (“ream and run”) can produce outcomes that are similar to TSA. However, these results have not been widely replicated and the indications for this procedure are not well defined.
Total shoulder arthroplasty – Surgical technique
Patient is placed into a modified beach chair position with two folded blue towels placed beneath the scapula to protract the scapula and bring the glenoid forward. The patient should be brought to the edge of the operating room table to ensure that the operative shoulder can be easily manipulated.
Surface anatomy is drawn out including the coracoid process with the planned incision drawn out from the medial edge of the coracoid to a point two fingerbreadths lateral to the top of the axially fold.
With the operative arm on a padded mayo stand the incision is made. The deltopectoral approach is utilized and soft tissue dissection is taken down until the cephalic vein is identified. The cephalic vein is taken medially while cauterizing identified braches to the deltoid.
The deltopectoral interval is developed with planes created deep to the pectoralis major bluntly and the deltoid using a cetalot elevator allowing the Koebel retractor to be inserted (medium size blade inserted medially and the small size blade inserted laterally).
A Hohmann retractor is then inserted superiorly above the coracoacromial ligament. A bovie is then used to outline the lateral border of the conjoint tendon. The plane underneath the conjoint tendon is developed bluntly and the Koebel blade is replaced deep to the conjoint tendon.
The fascia overlying the subscapularis tendon is exposed. Small vessels including the three sisters can be cauterized. The subscapularis is tenotomized from the lesser tuberosity, leaving a small cuff of tissue attached to allow for repair at the conclusion of the procedure. Be sure to hug the anatomic neck of the humeral head when taking down the inferior portion of the subscapularis and externally rotate the humerus to aid in exposure. Traction sutures are passed through the tenotomized subscapularis for later repair.
With the humeral head exposed any identified osteophytes can and should be removed with a rongeur. The long head of the biceps is cut and the axillary nerve is palpated inferiorly to identify its location.
Take the operative arm off the padded mayo stand and dislocate the humeral head with external rotation, extension and adduction.
Finish removing osteophytes and locate the anatomic neck.
A Hohmann retractor is placed between the rotator cuff and humeral head. Inspect the integrity of the rotator cuff and perform a repair if necessary.
Check version of the humeral head relative to the forearm.
Place a medium Darach retractor along the inferior anatomic neck to protect the axillary nerve.
Mark out the planned humeral head cut using the template provided with the arthroplasty set and cut along the anatomic neck with an oscillating saw.
Using a starting awl, obtain access to the medullary canal.
Ream until cortical scratch is felt making sure to stay lateral to avoid varus positioning.
Broach up to the appropriate size – leaving final broach in place.
Using a curved mayo scissors to dissect along the posteroinferior border of the subscapularis creating plane between the subscapularis anteriorly and the capsule posteriorly — allowing the capsule to be seen as a separate layer from the subscapularis and the fatty tissue containing the nerve inferiorly. The capsule can then be cut until the triceps is visualized.
With a fukuda retractor in place retracting the prepared proximal humerus posteriorly the glenoid is exposed.
Mark the glenoid vertically and horizontally along the equator. Allows for evaluation of any posterior glenoid wear that may be present. If necessary a bone graft wedge can be obtained from the previously osteotomized humeral head and used if necessary.
Use glenoid guide to drill a hole in the center of the glenoid. The glenoid is then reamed until punctate bleeding cancellous bone is visualized.
Two additional drill holes can be made around the central drill hold to ease keel creation. The keel impactor is then used.
Irrigate and pack the prepared glenoid with an epinephrine soaked sponge. Dry the glenoid. Cement in the glenoid component and hold the implant in proper position with the glenoid impactor until the cement hardens.
Humeral implant insertion
Remove the trial broach. Irrigate the canal.
Seat the final implant.
Insert trial heads – checking stability and range of motion (40 degrees of external rotation with the arm at neutral, 50% posterior translation of humeral component on glenoid, 60 degrees of internal rotation with the arm abducted 90 degrees).
Insert final head implant.
Use a minimum of 8 sutures to repair the subscapularis – 1st suture placed at the superior corner then continue inferiorly.
Tenodese the long head of the biceps tendon.
Close the deltopectoral interval with interrupted sutures.
2-0 vicryl suture for subcutaneous tissue reapproximation.
3-0 monocryl for the skin.
Pearls and Pitfalls of Technique
Pre-operatively, must discuss with the patient regarding unpredictable pain relief and the possibility of disease progression that can require additional future surgeries.
Application should be limited to mild disease – studies suggest unlikely to benefit if <2mm joint space remaining or severe disease on both humeral head and glenoid surface.
Typically combined with other procedures that address concomitant pain generators about the shoulder joint.
Although 30 degrees of flexion, internal rotation, and abduction has been recommended, optimal positioning of the fusion must be individualized for each patient.
Position of fusion must be confirmed intraoperatively so that the arm can be functional for feeding and hygiene.
Opposition and compression of bony surfaces (glenoid to humeral head / acromion to humeral head) are necessary for a successful union.
Post-operative occupational therapy to maximize function is essential.
Primary advantage is maintenance of bone stock for possible future revision.
Pre-operatively, must discuss with the patient regarding unpredictable pain relief and the possibility of disease progression that can require additional future surgeries.
Exposure of the glenoid can be technically challenging as majority of the humeral head is maintained.
For optimal exposure, the patient must be positioned to allow full adduction and extension of the shoulder.
Pre-operative CT scan can be valuable to clearly assess the glenoid anatomy and version.
Subscapularis tendon can be released via tenotomy or lesser tuberosity osteotomy. Regardless of the technique, subscapularis must be repaired with solid fixation.
Irreparable rotator cuff tear is a contraindication for TSA.
A rare, but severe, complication that may occur during shoulder surgery is nerve damage. In fact, the entire brachial plexus is at risk for injury during shoulder arthroplasty as the arm is often maximally extended and externally rotated for prolonged periods. This position has been documented to cause traction type neuropraxia, and can cause clinical symptoms post-operatively. Other nerves that are susceptible to injury include the musculocutaneous nerve just medial to the conjoint tendon, and the axillary nerve just inferior to the subscapularis and the inferior glenoid margin. Typically, increased awareness and careful dissection in these regions are sufficient to avoid injury to these nerves. However, some may choose to dissect and visualize one or both of these nerves during the surgery.
If post-operative nerve injury is observed, electrodiagnostic tests are recommended to document the extent of the injury and to establish a baseline of the injury. Unless direct transection of the nerve is suspected, most are neuropraxia type of injuries that will recover over 3-6 months. If clinical recovery is not evident after this period of observation, another electrodiagnostic examination should be obtained. If these studies also fail to document any evidence of recovery, surgical exploration with possible nerve repair or grafting may be required.
Another serious complication, especially for those with shoulder arthroplasty, is deep infection. Therefore, deep infection must always be considered if patients present with persistent pain, weakness, or hardware complication after their initial recovery. For most, documented infection will require surgical debridement with possible exchange or removal of implants. Principles of surgical debridement of infected arthroplasty are obtaining cultures and tissues prior to initiating antibiotic therapy, complete debridement of necrotic tissues including bone, copious irrigation, and consideration for use of antibiotic-impregnated cement spacer as needed. Given the frequency of propionibacterium acnes to infect the shoulder joint, its slow growth, intra-operative cultures must be maintained for at least 2 weeks before negative growth can be documented. Once the causative organism is isolated, prolonged parenteral antibiotic treatment will be required. If an organism cannot be isolated, broad spectrum antibiotic may be needed. Eradication of infection can be supported by serial laboratory examinations with C-reactive protein and Erythrocyte Sedimentation Rate. However, it must be noted that these laboratory examinations support, but do not define, the eradication of infection. Only after infection has been clinically eradicated, a revision arthroplasty can be considered.
Other general complications inherent to shoulder surgery include those related to the use of a general or regional anesthesia, patient positioning, wound complications, vascular injury, and intra-operative fractures. However, most of these complications are rare and can be minimized with careful pre-operative planning and meticulous surgical technique.
Post-operative rehabilitation after treatment for glenohumeral OA generally follows the sequence of 1) protection of the repaired construct, 2) restoration of shoulder motion, and 3) conditioning of the peri-scapular and shoulder musculature. For patients who were treated with arthroscopic debridement, there is often no repaired construct to protect and thus, these patients may initiate these protocols immediately with no restrictions. Patients with shoulder arthroplasty, on the other hand, will require initial period of protection for the repaired subscapularis (6 weeks of no active internal rotation). Therefore, these patients are often immobilized in a sling for a period of time and their initial motion exercises are performed within the limits allowed by the strength of the subscapularis repair.
After the repaired construct is allowed to heal, aggressive motion exercises must be instituted. Although strengthening exercises can also be started earlier in the rehabilitation, motion exercises may need to continue simultaneously until full motion is restored. In addition, strengthening exercises should also include the peri-scapular musculature for optimization of the outcome.
Outcomes/Evidence in the Literature
Izquierdo, R. “AAOS Clinical Practice Guideline Summary: Treatment of Glenhumeral Osteoarthritis”. JAAOS. vol. 18. 2010. pp. 375-382.
Bryant, D, Litchfield, R, Sandow, M, Gartsman, G, Guyatt, G, Kirkley, A. “A comparison of pain, strength, range of motion, and functional outcomes after hemiarthroplasty and total shoulder arthroplasty in patients with osteoarthritis of the shoulder. A systematic review and meta-analysis”. JBJS Am. vol. 87. 2005. pp. 1947-56. (Systematic Review of Literature: Primary total shoulder arthroplasty has been shown in a number of studies to have superior outcomes to hemiarthroplasty for those with osteoarthritis of the glenohumeral joint and an intact rotator cuff.)
Sandow, M, David, H, Bentall, SJ. “Hemiarthroplasty vs total shoulder replacement for cuff intact osteoarthritis: how do they fare after a decade?”. JSES. vol. 22. 2013. pp. 877-85.
Levine, WN, Fischer, CT, Nguyen, D, Flatow, EL, Ahmad, CS, Bigliani, LU. “Long-term follow-up of shoulder hemiarthroplasty for glenohumeral osteoarthritis”. vol. 94. 2012. (Level IV study: Long term follow-up of hemiarthroplasties for glenohumeral OA at average of 17 years post-op, show only a quarter of patients satisfied with their outcomes (Neer, EuroQol measures).)
Van Thiel, GS. “Retrospective analysis of arthroscopic management of glenohumeral degenerative disease”. Arthroscopy. vol. 26. 2010. pp. 1451-5. (Level IV study: Grade 4 glenohumeral OA, joint space of <2mm and large osteophytes are poor prognostic factors for pain relief after arthroscopic debridement for a degenerate shoulder.)
Millett, PJ, Horan, MP, Pennock, AT, Rios, D. “Comprehensive Arthroscopic Management (CAM) procedure: clinical results of a joint-preserving arthroscopic treatment for young, active patients with advanced shoulder osteoarthritis”. Arthroscopy. vol. 29. 2013. pp. 440-8. (Level IV study: Arthroscopic glenohumeral chondroplasty, removal of loose bodies and osteophytes with biceps tenodesis (‘CAM’ procedure for comprehensive arthroscopic management) shown to be a viable option for delaying need for arthroplasty in young high demand patients with moderate to severe osteoarthritis.)
Richards, RR, Beaton, D, Hudson, AR. “Shoulder arthrodesis with plate fixation: Functional outcome analysis”. JSES. vol. 2. 1993. pp. 225-39. (Outcomes of arthrodesis for osteoarthritis are better than those who receive arthrodesis for instability. Most common complication of arthrodesis for OA is pseudoarthrosis/nonunion while most common complications of arthrodesis for instability is continued sensation of instability.)
Kelley, JD, Norris, TR. “Decision making in glenohumeral arthroplasty”. J Arthroplasty. vol. 18. 2011. pp. 75-82.
Tibbetts, RM, Wirth, MA. “Shoulder arthroplasty for the young, active patient”. Instruc Course Lectures. vol. 60. 2011. pp. 99-104.
Litchfield, RB, McKee, MD, Balyk, R, Mandel, S, Holtby, R, Hollinshead, R. “Cemented versus uncemented fixation of humeral components in total shoulder arthroplasty for osteoarthritis of the shoulder: a prospective, randomized, double-blind clinical trial-A JOINTs Canada Project”. JSES. vol. 20. 2011. pp. 529-36. (Level 1 study: cemented humeral component for total shoulder arthroplasty gives better outcomes with respect to strength, range of motion and quality of life compared to uncemented.)
Glenohumeral osteoarthritis is a common pathology heralded by activity related pain, stiffness and limited function. Once conservative management options have been exhausted surgical intervention is warranted. While a number of options are available, total shoulder arthroplasty remains the most reliable treatment method for pain relief and functional improvement.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- The Problem
- Clinical Presentation
- Diagnostic Workup
- Non–Operative Management
- Indications for Surgery
- Surgical Technique
- Pearls and Pitfalls of Technique
- Potential Complications
- Post–operative Rehabilitation
- Outcomes/Evidence in the Literature