Total Shoulder Arthroplasty in Primary Osteoarthritis: Technique and Results



Total shoulder replacement in the setting of primary osteoarthritis has evolved over the years. Current techniques and modern implants allow for predictable pain relief and an improvement in overall function and patient satisfaction. Surgical technique, including principles of soft tissue balancing, and postoperative care following primary total shoulder replacement are detailed in this chapter.


  • 1

    An extensive preoperative evaluation, including a history and physical examination, as well as adequate imaging are essential for all patients.

  • 2

    Accurate preoperative planning will improve results and reduce complications.


  • 1

    Subscapularis management is determined by the amount of internal rotation contracture present preoperatively.

  • 2

    Adequate glenoid exposure is dependent upon muscular paralysis, adequate humeral bone resection, proper arm positioning, appropriate soft tissue releases, and proper retractors.

  • 3

    Soft tissue balancing is essential for adequate postoperative range of motion, function, and implant stability.


  • 1

    Posterior glenoid deficiency is common in the setting of osteoarthritis and must be addressed at the time of surgery.

  • 2

    Avoid overstuffing the joint in an effort to “improve” stability.


Not applicable.

The first prosthetic replacement of the glenohumeral joint dates back to 1893 and was performed by a French surgeon, Péan, who performed the operation in a patient with tuberculosis of the shoulder. The early prosthesis used by Péan was constrained and subsequently failed 2 years after implantation secondary to recurrent infection. In the early 1950s the modern era of shoulder replacements began. These implants were unconstrained and anatomically designed to replicate the shape of the humeral head. The first reported results after glenohumeral replacement for arthritis in a large series of patients was by Neer in 1974. All but one of the patients in this series underwent humeral hemiarthroplasty, with the remaining patient receiving a cemented all-polyethylene glenoid component. Of the 48 patients, 42 achieved a satisfactory result. Over the next decade, several authors reported on the implantation of a polyethylene glenoid component for the humeral head prosthesis to articulate with, which began the modern era of total shoulder arthroplasty.

The functional results of total shoulder replacement are variable and dependent upon the disease process for which the replacement is indicated. Unconstrained prosthetic implants have the most success when used for conditions characterized by an intact or reparable rotator cuff. Conversely, when the rotator cuff is absent or nonreparable, the results are much less satisfactory. The topic of this chapter is total shoulder arthroplasty in primary osteoarthritis; therefore, our discussion will assume that the arthroplasty is performed in the setting of an intact or reparable rotator cuff. In addition to discussing indications and contraindications, this chapter will address the pathologic anatomy associated with osteoarthritis, review the preoperative evaluation, discuss the surgical technique and the postoperative rehabilitative program, and lastly report on the results of the procedure.


The primary indication for total shoulder arthroplasty (TSA) in the setting of osteoarthritis is intractable pain that has failed to respond to conservative management and is causing a degree of functional disability that is unacceptable to the patient. Primary osteoarthritis is characterized by posterior glenoid wear or bone loss associated with anterior soft tissue contracture. Therefore, one must make a decision regarding TSA versus hemiarthroplasty. The indications for glenoid replacement depend on patient age and activity level. In older patients in whom glenoid longevity is less of a concern, TSA is the procedure of choice. Younger or more active patients can present a treatment conundrum because prosthetic replacement puts these patients at a high risk for multiple revisions. Other surgical options, such as debridement, interposition arthroplasty, and resurfacing arthroplasty, should be considered as alternatives to traditional hemiarthroplasty or total shoulder arthroplasty in these patients. These options can provide temporary symptomatic relief and functional improvement while preserving bone stock. However, there is a paucity of information in the literature regarding the success rate of these procedures. Generally speaking, we perform TSA in patients 50 years and older, whereas hemiarthroplasty is considered for patients younger than 50 and older than 40. For patients 40 years and younger, hemiarthroplasty is usually augmented with interposition allografts such as fascia lata, Achilles tendon, or lateral meniscus. These patient ages are obviously guidelines and should be considered in addition to glenoid bone stock, degree of humeral head collapse, activity level, and overall medical condition.


The only absolute contraindication to TSA is infection. Other relative contraindications include any relevant medical or physiologic issues that would compromise a patient undergoing the procedure and the necessary rehabilitation, such as nerve injury or severe medical comorbidities. Patients with a remote history of infection should be properly screened to rule out current infection. We recommend obtaining a complete blood cell count (CBC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, and cultures when indicated.

Excessive bone deformities of the posterior glenoid are rare in patients with primary osteoarthritis; however, in the rare instances when even with bone grafting the resulting glenoid vault has a depth of only up to 1 cm, the use of bone grafting should be avoided. In theses instances, we recommend hemiarthroplasty with or without interposition allograft. Another relative contraindication to glenoid resurfacing is excessive posterior laxity and/or subluxation that cannot be corrected. Such instability can lead to eccentric wear and ultimately mechanical failure.


Osteoarthritis in any joint is characterized by progressive joint space narrowing, subchondral sclerosis, and prominent osteophyte formation. Neer described the primary features of glenohumeral osteoarthritis to be thinning of the humeral articular cartilage, humeral subchondral cysts located slightly above the equator of the articular surface, and the presence of inferior humeral osteophytes, which are commonly seen on plain radiographs.

Glenoid retroversion is frequently increased in primary osteoarthritis, often secondary to asymmetric posterior wear, but can also be the result of congenital hypoplasia or developmental abnormalities. In addition, posterior humeral subluxation can be found in up to 45% of patients with primary osteoarthritis. Walch and colleagues et al. have classified glenoid bone deformity according to posterior bone wear, erosion, or overt bone loss. These include type A, concentric, with no humeral head subluxation (59%); type B, asymmetric wear of the posterior glenoid rim with posterior humeral subluxation (32%); and type C, glenoid retroversion of greater than 25 degrees with posterior humeral subluxation (9%) ( Fig. 8–1 ).

Figure 8–1

The five different morphologic types of the glenoid in primary osteoarthritis as described by Walch et al.

(Adapted with permission from Walch G, Badet R, Boulahia A, et al: Morphologic study of the glenoid in primary glenohumeral osteoarthritis. J Arthroplasty 14(6):756–760, 1999.)

Soft tissue insufficiency can also be a significant component in primary osteoarthritis. The most common soft tissue contracture involves the anterior capsule and, in severe cases, the subscapularis. Excessive posterior subluxation secondary to posterior laxity can be an issue, but in many cases the humeral head is displaced medially as well as posteriorly, resulting in less laxity than expected. Rotator cuff insufficiency is rarely a significant issue in the setting of primary osteoarthritis. Soft tissue balancing is an important component of TSA because eccentric loading of the glenoid component can lead to premature failure.

The degree of posterior glenoid deformity is a critical factor in glenoid replacement and determines which technique is used to correct the deformity and create a neutral glenoid surface. Therefore, it is essential that a thorough preoperative clinical evaluation be performed and adequate radiologic studies be obtained. A more thorough description of the various techniques to correct varying degrees of glenoid bone loss is discussed later in this chapter.



Patients who suffer from primary osteoarthritis of the shoulder typically present with an insidious onset of pain, which has gotten progressively worse over time, leading to a gradual loss of function. Many patients also complain of progressive stiffness associated with the discomfort. Typically the stiffness is most notable after a period of immobilization and improves with use.

Many patients note that the progressive loss of motion of their shoulder has led to an impairment of their activities of daily living; women may complain of difficulty fastening a brassiere or men may note difficulty reaching their wallet secondary to limited internal rotation. It is important to ask patients how their condition has interfered with their daily activities because the answers to such questions can provide insight into the degree of pain and disability that they are experiencing. Additionally, understanding patients’ occupation, hobbies, and desired level of activity can help the physician determine the appropriate level of surgical intervention. Finally, documenting prior treatment regimens that have or have not been successful also gives information about the disease course and severity.

It is important to be aware of the patients’ comorbidities not only for presurgical screening but also as a means of evaluating other conditions that may be a source of disability and limit rehabilitation.

Physical Examination

A thorough examination of the affected shoulder should be performed with particular attention to range of motion. As previously mentioned, osteoarthritis typically leads to a global loss of shoulder motion, particularly external rotation. Any internal rotation contracture must be noted because this dictates how the subscapularis is released at the time of surgery. Active and passive range of motion, as well as rotator cuff strength, should be documented and compared with the contralateral side. Painful crepitus associated with motion is common. The acromioclavicular (AC) joint should be palpated because an arthritic AC joint can be a pain generator and can be addressed at the time of surgery if needed.

Imaging Studies

Plain radiographs remain the most important radiographic study required in the diagnosis of osteoarthritis as well as for surgical planning. We routinely obtain a standard set of radiographs (anteroposterior [AP], transcapular lateral, and axillary lateral), and each radiograph provides different information that is crucial for preoperative preparation. The AP view is often done in internal and external rotation and allows for assessment of bone quality, identification of inferior osteophytes, and diameter of the humeral canal. The axillary lateral radiograph is crucial because it allows for evaluation of posterior glenoid wear and associated posterior subluxation ( Fig. 8–2 ).

Figure 8–2

Axillary lateral radiograph demonstrating decreased joint space but no excessive posterior subluxation or glenoid wear.

Computed tomography (CT) can provide a more definitive assessment of glenoid version and bone stock. CT scanning also allows for a more accurate determination of whether glenoid replacement is feasible and whether bone grafting is necessary. A study by Walch et al. reported that 45% of patients with primary osteoarthritis awaiting shoulder arthroplasty had posterior subluxation, and the average glenoid retroversion in this population was found to be 15.4 degrees (normal 1 to 2 degrees). We typically obtain a CT scan in all patients with external rotation of 30 degrees or less because posterior glenoid erosion and subluxation are common in patients with advanced internal rotation contracture.

Magnetic resonance imaging (MRI) can be helpful in patients with a suspected rotator cuff tear, but it is rare for patients with primary osteoarthritis to have significant rotator cuff pathology. We reserve MRI for patients with a history of rotator cuff repair or when superior escape is evident on plain radiographs ( Fig. 8–3 ).

Figure 8–3

MRI of a right shoulder demonstrating an intact subscapularis tendon (white arrow). We use MRI to evaluate the rotator cuff if a history or physical examination finding is consistent with cuff injury and/or if radiographic evidence of superior escape is present.


Stemmed humeral head replacements remain the most popular implant for shoulder replacement in the setting of primary osteoarthritis. The other option is humeral head resurfacing without a stemmed implant. The advantages of traditional stemmed implants include an easier exposure of the glenoid for resurfacing, greater familiarity, and a longer track record. However, stemmed implants also require greater bone removal and potentially a larger, more extensive surgical exposure. Humeral head resurfacing can be an option in younger patients where preservation of bone stock is critical and in patients without glenoid arthrosis. Although it is possible to resurface the glenoid without removing the humeral head, it can be very difficult compared with traditional humeral head removing implants. There are very few reported series regarding humeral head resurfacing. Levy and Copeland reported good results of patients undergoing humeral head resurfacing with a 5- to 10-year follow-up. They also reported that patients with primary osteoarthritis had the best outcome and that only 8% required revision during the follow-up period. Recently, we (DMD) have been using a “ministem” humeral component with much success. We believe this system allows for excellent clinical results as well as the benefit of potentially less postoperative blood loss and pain. We hope to report on these results in the near future.

Numerous glenoid replacement prostheses are available today for shoulder arthroplasty. These include all polyethylene designs, metal-backed designs, and hybrid designs with metal peg sleeves but no metal backing. There is also some variety with regard to articular conformity and constraint. Some studies have suggested that the arthroplasties characterized by nonconforming radii of curvature yield more physiologic translations and exhibit lower loosening scores than conforming designs. However, more studies are needed to confirm this.


In this section we discuss the authors’ preferred type of anesthesia, patient positioning, surgical approach, and surgical technique, as well as implant considerations for patients with primary osteoarthritis undergoing TSA. Additionally, we discuss some specific techniques we use when we encounter some of the unique bony pathology associated with osteoarthritis.

Anesthesia And Patient Positioning

Options for anesthesia in the setting of shoulder arthroplasty include general anesthesia, regional anesthesia (e.g., interscalene block), or a combination of both. Each one of these options has advantages and disadvantages, and each patient should have a thorough discussion with the anesthesiologist before deciding which type of anesthesia to receive. At one of our institutions, (Hospital for Special surgery—DMD, CCD), all patients undergoing shoulder arthroplasty receive an intrascalene block. This is our preferred method of anesthesia because it is safe and provides muscle relaxation as well as intraoperative and postoperative pain relief to facilitate early postoperative motion. However, it should be noted that some authors have reported an increase in induction time and increased risk of neurologic, pulmonary, and vascular complications associated with an intrascalene block. Using a combination of general anesthesia and an intrascalene block potentially provides for the advantages of both techniques and is often used.

Following induction of anesthesia, the patient is placed in the beach-chair position with the torso at approximately 45 degrees from horizontal ( Fig. 8–4 ). The head and body should be stabilized and all bony prominences well padded. The affected extremity should be draped free and mobile enough to be placed into a fully extended and adducted position during the procedure. Once the patient is adequately positioned, a final measurement of external rotation is noted and the patient is prepped and draped in the usual sterile fashion.

Figure 8–4

Intraoperative photograph of a patient in the beach-chair position in preparation for a right total shoulder replacement.

Preferred Approach

Several surgical exposures have been described for shoulder replacement. The deltoid-reflecting approaches offer superior exposure at the expense of potential deltoid morbidity or osteotomy nonunion. In most cases of primary osteoarthritis, every reconstructive situation encountered can be adequately performed through an anterior deltopectoral approach without detaching the deltoid origin or insertion. Therefore, this remains our preferred approach in this patient population.

Incision And Exposure

Superficial Exposure

The deltopectoral approach is begun by making a 10- to 15-cm incision beginning at the clavicle just medial and superior to the coracoid and extending distally along the deltopectoral groove ( Fig. 8–4 ). The subcutaneous fat is then split in line with the incision, which allows for identification of the cephalic vein. The cephalic vein is a landmark of the deltopectoral interval and is a large draining tributary for the upper extremity. We prefer to preserve the cephalic vein, and its preservation may be especially important in patients with potential venous outflow difficulties (e.g., history of mastectomy, radiation, etc.).

The vein may be retraced medially or laterally; some have proposed that the vein be retracted laterally to reduce to preserve branches to the deltoid, whereas others believe the vein should be retracted medially to avoid excessive tension on the vein itself. We typically retract the vein laterally but recognize that retracting it medially is probably fine as long as it is handled carefully and copious hemostasis achieved.

The first step in gaining adequate exposure is freeing the deltoid from the pectoralis major, starting at the clavicle and extending to the deltoid insertion on the humeral shaft. The deltoid is then retracted laterally, and the interval between the surgical neck of the humerus and the deltoid is then bluntly dissected using an elevator. It is important to keep the elevator on bone because the axillary nerve crosses the deep deltoid in this region.

The tip of the coracoid process is then identified, and a plane is carefully developed between the strap muscles and the overlying pectoralis major muscle. The tendinous insertion of the pectoralis major on the humerus is then identified, and, if needed, the upper 1 cm can be released to improve exposure ( Fig. 8–5 ). A self-retaining retractor, such as the Tieman, is then placed with one limb under the deltoid laterally and one limb under the pectoralis muscle.

Mar 22, 2019 | Posted by in ORTHOPEDIC | Comments Off on Total Shoulder Arthroplasty in Primary Osteoarthritis: Technique and Results

Full access? Get Clinical Tree

Get Clinical Tree app for offline access