CHAPTER SYNOPSIS
Humeral head hemiarthroplasty is a proven effective method of improving pain and functional deficits of the shoulder related to conditions including osteoarthritis, post-traumatic arthritis, humeral head osteonecrosis, rheumatoid arthritis and cuff tear arthropathy when conservative measures fail. The use of a prosthetic glenoid component, which is responsible for the majority of failures of total shoulder arthroplasties, is avoided. However, it remains critically important that the glenoid be adequately assessed and that nonprosthetic treatment of glenoid pathology be performed as necessary.
IMPORTANT POINTS
The most important issue to address is recognition of the appropriate indications for humeral head hemiarthroplasty. Hemiarthroplasty can be utilized when there is a normal glenoid, a concentric glenoid consisting of eburnated bone, or a nonconcentric glenoid that can be converted to a smooth concentric surface by reaming. The humeral head must also be centered in the glenoid preoperatively or at the time of surgery by means of soft tissue balancing and glenoid preparation. When glenoid reaming is not required, the labrum may be maintained circumferentially.
CLINICAL/SURGICAL PEARLS
- 1
A hemiarthroplasty is contraindicated in the presence of a pitted or eroded osteopenic glenoid or a nonconcentric glenoid not correctable by reaming.
- 2
Soft tissue balancing, including anterior and inferior capsular release and circumferential mobilization of the subscapularis tendon, is necessary for optimal results.
- 3
Erosion of the natural glenoid leading to the need for conversion to total shoulder arthroplasty is a concern that must be better defined over the long term.
CLINICAL/SURGICAL PITFALLS
- 1
If inadequate glenoid anatomy exists, conversion to a total shoulder should be considered.
- 2
An osteotomy of the humeral head that is too lateral or in excessive retroversion will compromise the rotator cuff insertion. The plane of the osteotomy should exit superiorly just medial to the greater tuberosity, and version must be controlled to avoid removing the posterior cuff attachments. Enhanced exposure can help to avoid this mistake.
HISTORY
In 1953, Neer described the first use of a metallic hemiarthroplasty for the treatment of proximal humeral fractures. Given the success of this implant for trauma, the use of this prosthesis was extended to other diagnoses, including humeral head osteonecrosis, rheumatoid arthritis, and osteoarthritis. Although significant success was encountered with this procedure, poorer outcomes were encountered in the presence of incongruent and eroded glenoids. Neer later presented the results of his first design of a polyethylene glenoid prosthesis as part of a total shoulder arthroplasty. A number of other implant designs were introduced with and without glenoid replacement. Many difficulties were encountered in achieving glenoid fixation, particularly with fully constrained ball-and-socket implants and partially constrained implants with hooded glenoid components. Bipolar implants and silicone and acrylic implants have also been described. The most reliable results have been produced in series utilizing humeral head hemiarthroplasty or unconstrained total shoulder arthroplasty implants.
As this brief history of shoulder prosthetic design illustrates, the ideal shoulder prosthesis has yet to be developed and likely would take different forms for varying pathologic conditions. Important evolution has taken place in the design of the humeral component. However, the greatest controversy has tended to focus on the glenoid, and this controversy remains active today. Because most failures of total shoulder arthroplasties occur at the glenoid, treatment options that avoid prosthetic replacement of the glenoid can have definite advantages. It is important to remember continuously however, that the decision to avoid prosthetic replacement of the glenoid is not in any way equivalent to ignoring or failing to address important glenoid pathology.
Statistics on shoulder arthroplasty procedures performed in the United States are much less detailed than those available for hip and knee arthroplasty. In addition, the results of a very small percentage of these are available for review in the published literature. Most studies arise from centers performing a large volume of shoulder arthroplasties. Jain et al. demonstrated that the results of shoulder joint replacement can vary considerably based on the case volume in a surgeon’s practice. Aspects of surgical technique that lie on steep learning curves include exposure of the glenoid, release of soft tissue contractures about the shoulder, and addressing eccentric glenoid wear. These surgical steps are of equal and perhaps greater importance when a prosthetic glenoid component is not utilized. Therefore, humeral head hemiarthroplasty must not be seen as a means of avoiding any of these critical technical aspects.
The term humeral head hemiarthroplasty refers to the treatment of glenohumeral joint pathology with prosthetic replacement only of the humerus. Several types of humeral head replacement are currently available. Partial articular surface replacements involve insetting a metallic component into a focal area of cartilage loss on the humeral head. The remainder of the head and the entire glenoid are left with their native articular surfaces intact. Humeral head resurfacing involves reaming of the entire humeral head without resection and fixation of a metallic component overlay. The native glenoid is usually left intact. These two prosthetic options are addressed in Chapter 4 , Chapter 5 , respectively. This chapter deals specifically with techniques that involve resection of the humeral head and insertion of a metallic stem into the medullary canal. Depending on the pathologic condition and surgeon preference, either a humeral head aimed at restoring normal anatomy or one providing extended superior coverage over the region of the greater tuberosity can be used. Bipolar implants have been used in the past but currently are employed infrequently due to functional concerns of overstuffing the shoulder joint with the oversized humeral head. This has led to excessive distension of the capsule, decreased range of motion, and pain resulting from increased joint reaction force on the native glenoid.
With the decision to pursue nonprosthetic treatment of the glenoid in shoulder arthroplasty, several treatment choices remain. The first option is to leave the native glenoid intact. If the surface is irregular, eccentrically worn, or improperly oriented, the glenoid may be reamed to a smooth concavity without prosthetic implantation. Interposition of the anterior capsule has been reported by Burkhead for revision cases as well as primary hemiarthroplasty in young, active patients. This topic is addressed in detail in Chapter 7 . Partial glenoid defects can also be treated with osteotomy or augmentation with bone grafts.
Engelbrecht and Heinert reported on a technique of structural bone grafting a deficiency of the superior glenoid to act as a buttress against load from a superomedially migrated humeral head in cuff-tear arthropathy. The resected humeral head typically provides a convenient source of autograft in primary arthroplasty procedures.
A variety of soft tissue procedures may also be indicated as part of the hemiarthroplasty procedure. These include the release of the motion plane between the undersurface of the deltoid and the humerus and rotator cuff. A circumferential release of the subscapularis and excision of the anterior capsule is routinely required. The anterior capsule is often contracted and requires release. Significant rotator cuff tears are not common in primary osteoarthritis but can often be encountered in other pathologic conditions such as posttraumatic and inflammatory arthritis. Sperling et al. and Wirth et al. demonstrated that the range of motion and strength achieved postoperatively was not negatively affected when comparing shoulders with a rotator cuff tear repaired at the time of hemiarthroplasty to shoulders with an intact cuff. Fraying of the long head biceps tendon may best be addressed with tenodesis and resection of the intra-articular portion. It is also of great importance to assess the integrity of the coracohumeral arch. Loss of the coracoacromial ligament along with a massive irreparable rotator cuff tear may lead to unbalanced pull from the anterior deltoid, pectoralis major, and other anterior muscle forces, leading to anterosuperior instability. This condition is associated with poor functional outcome and patient satisfaction. Published series of attempts at reconstruction of the arch have not demonstrated reliable success. However, reverse ball-and-socket total shoulder arthroplasty has shown great promise for this condition.
INDICATIONS AND CONTRAINDICATIONS
The primary indication for humeral head hemiarthroplasty, as indeed any arthroplasty, is for the amelioration of the pain of joint arthrosis resistant to conservative measures. It also aims to improve function in the patient’s activities of daily living. The patient must be of adequate health to allow the procedure to be performed safely and must be cooperative with the postoperative rehabilitation regime.
Beyond these subjective patient factors, the anatomic indications for shoulder hemiarthroplasty include either a concentric glenoid of eburnated bone or a nonconcentric glenoid that can be converted intraoperatively to a smooth concentric glenoid with acceptable version through reaming. The humeral head must also be well centered in the glenoid preoperatively. Alternatively, it must be possible to achieve intraoperative centering with soft tissue balancing and glenoid preparation.
Surgeon and implant factors provide important indications. Implants systems in common use provide a variety of curvature and thickness options that are important to restore normal anatomy and soft tissue balance. The surgeon must be proficient in techniques required to gain adequate exposure of the entire shoulder joint, including the glenoid. The surgeon must also be comfortable with techniques for releasing extra-articular and capsular contractures, addressing eccentric glenoid wear, and achieving soft tissue balance.
Relative contraindications include a history of remote infection of the glenohumeral joint, previous shoulder surgery, alcoholism, smoking, chronic narcotic use, substantial Parkinson disease, and neuropathic arthritis. These conditions are known to lessen the chances of a good result. Chronic weight-bearing on the upper extremities due to paraplegia or crutch use may also negatively affect outcome, though the longevity of prostheses in the long term has not been established. More definitive contraindications for humeral head hemiarthroplasty include a cystic or pitted osteopenic glenoid, because it will not provide an adequate bearing surface against the prosthetic head and will be prone to persistent pain and progressive erosion. A nonconcentric glenoid not correctible by reaming or moderate or severe glenohumeral subluxation that persists following glenoid preparation and attempts at soft tissue balancing are best served by total shoulder arthroplasty. The presence of active or recent infection, absent deltoid function, poor patient general health, and an inability or unwillingness to cooperate with a physician-directed rehabilitation program are further contraindications that often present few or no surgical options.
Indications for humeral hemiarthroplasty are particularly controversial for patients with inflammatory arthropathies. Rheumatoid patients have been reported to have rotator cuff tears in one third of arthroplasty cases. Many more patients likely present with attenuated or poorly functioning cuff tissue. These patients also nearly always suffer from osteopenia related to chronic corticosteroid use and the direct effects of their disease. As such, rheumatoid patients are at increased risk of loosening of a glenoid component due to suboptimal bony fixation and the “rocking horse” effect of edge loading by a nonconcentric humeral head. However, at the same time, the rheumatoid glenoid is typically soft, pitted, and centrally eroded. This does not represent the optimal bearing surface for a humeral head hemiarthroplasty. In a series of 19 rheumatoid patients, the hemiarthroplasty procedure gave good pain relief and improved function without the concern of early glenoid component failure. The status of the rotator cuff significantly affected results.
The indications for treatment of cuff-tear arthropathy with humeral head hemiarthroplasty follows the same subjective patient factors listed for anatomic head replacement in primary or secondary osteoarthritis. However, when assessing anatomic criteria, it is expected that there is superior erosion of the glenoid and that a concentric alignment of the humeral head and glenoid will not be achieved with the hemiarthroplasty procedure. An intact coracoacromial arch and anterior deltoid are important to provide a fulcrum for glenohumeral rotation, prevent anterosuperior dislocation, and provide power for shoulder elevation. Currently, the primary surgical options consist of standard hemiarthroplasty, hemiarthroplasty with use of an expanded-coverage humeral head, and reverse ball-and-socket total shoulder arthroplasty. Hemiarthroplasty is indicated for younger, active patients for whom the concern regarding longevity and revision of a reverse implant is an overriding concern. It is also indicated when there is secondary concentricity of the glenohumeral joint. Patients who have significant pain but have maintained good active elevation of the shoulder may also be indicated for a hemiarthroplasty. Contraindications again closely follow those discussed earlier, with particular attention to the presence of an intact coracoacromial arch to avoid anterosuperior escape of the humeral head.
SURGICAL TECHNIQUE
The surgical technique of humeral head hemiarthroplasty for treatment of osteoarthritis has recently been described in detail in the literature. Successful humeral head hemiarthroplasty begins with appropriate patient selection with the indications as discussed earlier. Preoperative planning is carried out to confirm that the patient is an appropriate candidate. Good-quality anteroposterior and axillary radiographs of the shoulder allow for identification of key parameters for successful replacement. The appropriate starting point for reaming is determined. The medullary canal is evaluated both for the maximal stem diameter that can be placed as well as the shape of the canal. A conical metaphysis will provide a good match for press fit of the prosthetic stem, whereas a cylindrical stem may need supplementation of bone graft or cement to achieve implant stability. Radiographs are also critical for identification of abnormal joint anatomy that will influence surgical technique. This can include curvature of the medullary canal or abnormal position of the tuberosities.
Anterior or posterior glenoid erosions are generally best seen on the axillary radiograph. The orientation of the glenoid face should be measured in relation to an axis line drawn down the body of the scapula. The normal glenoid possesses 0 to 10 degrees of retroversion. Abnormalities of version and contour, such as the biconcave glenoid typically seen with posterior erosion in osteoarthritis, must be addressed at surgery. Superior glenoid erosion associated with a high-riding humeral head in cuff-tear arthropathy is clearly seen on the anteroposterior radiograph. A CT scan can assist in evaluating these parameters and should be considered for surgical patients with significant limitations of elevation or external rotation or with bony abnormalities not fully defined by plain radiographs. Preoperative planning also includes a thorough discussion to ensure that the patient understands the critical role their participation in the postoperative rehabilitation plays in achieving excellent function following arthroplasty surgery.
Surgery is most often performed under general anesthesia, though regional anesthetic techniques such as interscalene blocks can be useful adjuncts for postoperative pain control. The patient is placed in the beach-chair position with the head elevated 30 degrees. The patient is positioned such that the surgeon has access to the entire shoulder girdle and the shoulder can be extended vertically and externally rotated during the case to expose the humeral head for resection and insertion of the prosthesis ( Fig. 6-1 ). Several operating table attachments are commercially available to assist in positioning. Of commonly available equipment, a McConnell headrest (McConnell Orthopedic Manufacturing, Greenville, Tex.) provides good head support while the patient is positioned at the edge of the operating table. There are also attachments that can assist in holding the forearm to free assistants from duty positioning the shoulder.
An extended deltopectoral approach is utilized, with careful preservation of the entire deltoid origin and insertion. Care is also taken in the use of self-retaining retractors to avoid damage to the deltoid muscle and cephalic vein. The subscapularis is elevated subperiosteally from the lesser tuberosity, starting just medial to the bicipittal groove in order to maximize tendon length. If the tendon has good thickness and strength, it can be elevated directly from the anterior capsule. The interval between subscapularis and capsule is most apparent medially and inferiorly where the muscle has not yet fully transitioned to tendon fibers. If the tendon is atrophied or friable, it is a good idea to leave the lateral aspect of the anterior capsule attached as reinforcement. The capsule must then be incised more medially and resected to allow release of the subscapularis excursion. With adequate protection of the axillary nerve, the inferior capsule should be released beyond the origin of the long head of the triceps on the glenoid. If the glenoid is to be preserved without reaming, capsular releases should be performed just beyond the labrum, which should be preserved. Adhesions from the base of the coracoid process are released, and the subscapularis muscle-tendon unit is released circumferentially about its long axis ( Fig. 6-2 ). Reattachment of the tendon is performed with grasping stitches of a heavy braided nonabsorbable suture such as 1-mm Cottony Dacron (Deknatel, Fall River, Mass.) passed through 1- to 2-cm-long bone tunnels drilled from the anatomic humeral neck to the anterior surface of the humeral head osteotomy ( Fig. 6-3 A ). This reattachment more medially produces a functional lengthening of the subscapularis to assist in soft tissue balance and motion restoration ( Fig. 6-3 B ). Once adequate soft tissue releases have been performed, the shoulder can be anteriorly dislocated with extension of the arm vertical to the floor. If there is excessive tension during this maneuver, further releases are likely to be required.