Arthritis or degenerative joint disease (DJD) in the shoulder of young athletes is a disabling condition that often leads to an end to sports participation and is extremely challenging to treat. Arthritis in the shoulder, defined as a degeneration of the articular cartilage of either the humeral head or glenoid, is caused by multiple etiologies, including primary osteoarthritis and secondary causes such as trauma, iatrogenic disease, and rheumatoid disease. Although each subtype has a unique set of considerations and treatment options, all have the same end goal of providing a functional, pain-free shoulder that has the durability to withstand the forces of daily and recreational activities for the duration of the athlete’s life—a particularly important and challenging consideration in this young, active population. These options encompass a wide range of modalities, including both operative and nonoperative treatments.
Shoulder DJD begins with loss of mobility and the inability to maintain function because of contracture and pain in and around the shoulder. This condition eventually leads to difficulty in continuing recreational activities, which is a substantial hurdle for persons who wish to lead an active lifestyle. A wide spectrum of disease can be present, from focal cartilaginous softening to diffuse, full-thickness lesions of both the glenoid and humeral head.
In the United States, the overall incidence of glenohumeral chondral lesions and their natural history are largely unknown. These lesions are often found incidentally at the time of arthroscopy for other pathologies. The causes of these lesions range from idiopathic to specific underlying pathologies that eventually lead to osteoarthritis. Chondrolysis is a specific form of osteoarthritis has been studied in national registries and has an incidence of 5.5 per 10 million person-years. It has an increased incidence in persons who have diabetes and in those who have undergone an orthopaedic procedure. Nakagawa et al. studied primary osteoarthritis, specifically excluding patients with rheumatoid disease and those with rotator cuff arthropathy, and found an incidence of 0.4% in persons with orthopaedic problems versus 4.6% in persons who presented with shoulder disease.
Primary osteoarthritis is typically found in persons older than 60 years. In the young, healthy, active population, the incidence would be expected to be much lower, although specific data are lacking in the current literature. Cadaveric studies have shown an increased incidence of chondral lesions associated with rotator cuff tears; in shoulders with tears of the rotator cuff, defects in the glenoid were found in 32% and humeral defects were found in 36% compared with rates of 6% and 7%, respectively, in shoulders without rotator cuff tears.
Numersous studies have looked at the incidence of chondral injuries in patients who have had a shoulder dislocation. In 1997 Taylor and Arciero reported that 57 of 63 patients had a Hill-Sachs lesion after anterior dislocation; 40% were chondral lesions, and 60% were osteochondral lesions. Furthermore, arthroscopic evaluation of 212 patients demonstrated a 23% incidence of glenoid defects and an 8% incidence of humeral head defects in persons who had one dislocation, whereas persons who reported more than two dislocations had a significantly increased incidence of arthritic lesions (27% and 36%, respectively).
Primary shoulder osteoarthritis is a rare phenomenon in the young athlete; this disease is found mostly in patients older than 60 years, and its etiology is unknown. It begins with stiffness of the glenohumeral joint, followed by joint space narrowing and humeral head and inferior glenoid osteophytes (which is considered the classic finding), with an intact rotator cuff. Samilson and Prieto first described radiographic parameters for glenohumeral arthritis in 1983. Since then the classification has been revised, and in 2002, Guyette et al. further refined this classification, with radiographs evaluated for humeral osteophytes, glenoid changes, and loss of joint space, collectively.
Walch et al. have further noted that fixed posterior humeral head subluxation may precede the development of osteoarthritic changes and may be the first sign of primary glenohumeral arthritis. Patients often have anterior capsular tightness that causes a static subluxation of the humeral head in relation to the glenoid. As the arthritis progresses, more osteophyte formation develops, along with progressive central loss of cartilage on both the humeral and glenoid sides of the joint.
Although the rates of shoulder arthritis have been demonstrated to be higher in patients who have sustained a shoulder dislocation (so-called dislocation arthropathy ) and in patients who have had that dislocation treated surgically (so-called capsulorrhaphy arthropathy ), we believe that these entities, although deserving of discrete consideration, may have significant overlap and may indeed be various points on a spectrum of the same disease process. However, to date, the literature has separated these processes into two distinct entities.
Impaction and shear across the glenohumeral joint causes an articular cartilage lesion during shoulder dislocation with an incidence of chondral and osteochondral lesions of 47% and 46%, respectively. The condition was initially described by Neer et al. in 1982, and Samilson and Prieto classified the condition in 1983. This form of arthritis is associated with increasing age at the initial event, direction of dislocation (with the posterior direction causing more changes than the anterior direction), and associated glenoid fractures. However, the number of dislocations and prior stabilization procedures were not associated with development of shoulder arthritis. Interestingly, Matsoukis et al. reported no difference between patients treated with and without surgery in regard to the severity of arthritis, and the overall incidence of arthritis after shoulder dislocation in patients treated nonoperatively has been reported to range from 10% to 20%. Most recently, 25-year follow-up data have been reported by Hovelius and Saeboe in 257 shoulders that sustained primary anterior dislocation. These investigators showed that shoulders were radiographically normal in only 44% of patients at 25 years and that the development of arthrosis was related to age at primary dislocation (>25 years), recurrence of instability, participation in high-energy sports causing injury, and a history of alcohol abuse. Although the rate of moderate/severe arthritis was 18% in shoulders without a recurrence compared with 26% for surgically stabilized shoulders, the rate of arthrosis in these surgically stabilized shoulders was less than in shoulders that were reported to have “stabilized over time,” raising a significant question regarding the role of recurrent instability in the future development of arthrosis.
Capsulorrhaphy arthropathy, as coined by Matsen and Wirth, is described as arthrosis that is attributed to overtightening of the capsular structures and results in abnormal translation of the humeral head opposite the direction of capsulorrhaphy. Usually found in male patients at a mean age of 45 years, it has been linked to length of time since follow-up, amount of external rotation contracture, and age at initial trauma. Biomechanical testing of selective capsular plication has shown that this mechanism causes alterations in humeral head translation. These translational differences lead to nonanatomic biomechanics, asymmetrical cartilage wear, and ultimately arthritis ( Fig. 56-1 ).
Surgical options have evolved as we have come to better understand the effects of biomechanics on shoulder articulation. Surgical options can now be divided into anatomic and nonanatomic procedures. Most nonanatomic repairs have been abandoned, including the Putti-Platt and Magnuson-Stack methods ; however, the Bristow-Latarjet method (although modified from its original description) has continued to be a mainstay of treatment in cases of anterior instability in the setting of bony glenoid deficiency.
In contrast to most nonanatomic repairs, anatomic procedures including Bankart repair and capsulorrhaphy have become the mainstay of treatment and have evolved significantly with use of arthroscopic techniques for primary and revision repairs for instability; some persons argue that these procedures are the current standard of care. However, this paradigm shift from primarily open to arthroscopic techniques leaves unanswered questions about the long-term outcomes of open versus arthroscopic procedures, especially with regard to arthritis.
Long-term data indicate the presence of degenerative changes after the Putti-Platt procedure in 30% to 61% of patients at 9 to 26 years of follow-up. These results have led to the abandonment of this procedure; however, the Bristow-Latarjet reconstruction remains part of the current treatment algorithm even though it has a reported incidence of arthritis of as high as 49% at 14 years. It may be argued that several technical details in the techniques used in these procedures (specifically, complete subscapularis takedown) are not currently performed, which may greatly affect outcomes, although few studies refuting this high rate of arthrosis have been reported. However, more recent study has demonstrated that despite an overall incidence of arthrosis in 34% of patients, moderate or severe arthrosis was present in only 6% (which correlated with worse outcomes), whereas the patients with only mild arthrosis had no degradation of outcomes from joint damage after the Bristow-Latarjet procedure. These studies have shown that the development of arthrosis may be more closely related to graft malposition than other factors previously thought to be responsible.
Chondrolysis of the glenohumeral joint is a devastating problem that has proven to be one of the most vexing diagnoses faced by shoulder surgeons during the past decade. It has many associated (although not well proven) causes and is characterized by unexpected pain and loss of joint mobility for weeks to months, typically after arthroscopic shoulder surgery. Global dissolution of cartilage occurs on both the glenoid and humeral head with the characteristic findings of joint space narrowing, periarticular bone edema, and subchondral cystic changes, although without osteophyte formation. Although this condition has been associated with the use of radiofrequency energy in the shoulder, low-grade infection, intraarticular injection of contrast medium, use of bioabsorbable suture anchors, and postoperative pain pump infusion, small case series and case reports have led to difficulty in extrapolating meaningful data with regard to the true etiology of this condition.
Recently, 375 patients who underwent an arthroscopic intraarticular shoulder procedure by an individual community orthopaedic surgeon were evaluated specifically with regard to chondrolysis. This cohort demonstrated a 13% incidence of chondrolysis, with all cases associated with an intraarticular postarthroscopic infusion of local anesthetic (bupivacaine or lidocaine); chondrolysis did not develop in any patients who underwent arthroscopy without a postoperative anesthetic infusion. Additionally, Bankart repair, arthroscopic debridement, suture anchor placement in the glenoid, and operative time were associated with this finding. This finding of the association between a postoperative anesthetic infusion and chondrolysis has been further supported by the results of a recent systematic review of 100 cases of chondrolysis in the literature, with this situation demonstrated in 59% of the cases reviewed.
Additionally, it has been demonstrated that the development of chondrolysis after a postoperative anesthetic infusion may be a dose-related phenomenon, which has been demonstrated in several basic science studies and supported in a recent clinical cohort. When a high-flow pain pump was used postoperatively, chondrolysis developed in 16 of 32 patients, whereas it developed in just 2 of 12 patients when a low-flow pump containing bupivacaine, epinephrine, and infusate was used.
With regard to low-grade infection as a causative agent, in their systematic review of 100 cases, Scheffel et al. reported that of 91 patients who underwent revision surgery, 41 had microbiologic cultures taken at the time of revision surgery. Growth was reported for only 3 of these 41 cases, with Propionibacterium acnes reported in all 3 cases, and no finding of any other positive cultures mentioned.
The possibility that bioabsorbable suture anchors are an etiologic agent has also recently been analyzed by Dhawan et al., who performed an evaluation of the 1,072,000 bioabsorbable anchors placed about the shoulder in 2008. Of these cases, 10 suture anchor–related complications were reported to the Food and Drug Administration, all of which were thought to be related to anchor malposition and typically demonstrated geographic rather than global cartilage loss. This finding was viewed as a best-case scenario, in that many cases of complications may not be reported; however, the authors still concluded that use of bioabsorbable anchors remains a safe practice when they are implanted properly based on the available literature.
Unfortunately, a reliable treatment algorithm or diagnostic modality to definitively identify patients with chondrolysis does not currently exist. Chondrolysis remains largely a diagnosis of exclusion, although surgeons need a high index of suspicion in the setting of failed previous surgery with early, progressive, and advanced arthritic changes of the glenohumeral joint. The outcomes of further definitive treatment of these patients have also been disappointing, and chondrolysis thus remains a major challenge for shoulder surgeons.
Rheumatoid arthritis is an inflammatory arthritic condition that affects the synovial linings of both small and large joints. Although other inflammatory arthritides may affect the young athletic population (particularly psoriatic arthritis, spondyloarthropathy, and reactive arthritis), most is known about rheumatoid disease in this setting. The inflammatory nature of the disease leads to a disabling secondary erosive arthritis. It most commonly affects the small joints of the body but is found in the shoulder in 90% of patients with chronic rheumatoid disease. Rheumatoid disease in the shoulder is progressive in nature with early symptoms of pain, swelling, and decreasing shoulder motion. With disease progression, extraarticular structures become involved, including the subacromial bursa, acromioclavicular (AC) joint, and rotator cuff. Radiographic progression begins with medial migration of the humeral head into the glenoid with characteristic central erosion of the articular surface ( Fig. 56-2 ). Bone quality is typically osteopenic with periarticular erosions involving the superior and medial humeral head. Cystic formation may occur at the rotator cuff insertion, which, coupled with intrinsic degeneration of the rotator cuff, may lead to functional rotator cuff deficiency, static superior migration, and uneven joint erosion. Late sequelae of the rheumatoid shoulder leads to painful joint destruction, loss of bone stock (which has significant implications for reconstructive options), rotator cuff compromise, and poor overall function. Further evaluation, especially in the preoperative setting, of all patients with rheumatoid arthritis should also include flexion and extension lateral views of the cervical spine to check for cervical instability and further causes of rheumatologic problems that may affect the shoulder joint concomitantly.
Osteonecrosis (also known as avascular necrosis or AVN) is another form of progressive glenohumeral arthritis that may be encountered during evaluation of the shoulder of a young athlete with arthritic disease. Osteonecrosis is characterized by the development of avascular regions of periarticular bone, resulting in infarction, necrosis, and ultimately collapse of the bony architecture of the subchondral plate with resultant deformity and arthrosis as a result of incongruity. Early workup of this disease should include testing for sickle cell anemia, particularly in the African American population. Also to be considered etiologically is protein C and S deficiency, factor V Leiden, and hyperlipidemia.
Although many causes of osteonecrosis have been described, including dysbarism, hemoglobinopathies, coagulopathies, Gaucher disease, and connective tissue disorders, steroid use and alcohol abuse predominate clinically. If radiographs reveal a crescent sign and collapse, the contralateral shoulder should be imaged to rule out asymptomatic bilateral disease, and use of the aggravating agent (if identified) should be stopped to prevent further progression of the disease. Collapse of the articular surface is secondary to fractures in weak subchondral bone and pain is often the most common complaint, specifically with difficulty sleeping and pain interfering with activities of daily living. Pain is usually present with flexed and abducted shoulder positions because of the incongruous central superior position of the humeral head contacting the glenoid in this position.
Classification of osteonecrosis of the humeral head has been described by Cruess ( Table 56-1 ). Osteonecrosis has six stages based on radiograph and magnetic resonance imaging (MRI); these stages are largely based on and parallel the original description in the femoral head by Arlet and Ficat. Indeed, osteonecrosis of the humeral head is similar to that in the femoral head, first with edema visible only on MRI, followed by sclerosis and then the development of the crescent sign, with further collapse of the articular surface and finally degeneration on both sides of the joint ( Fig. 56-3 ).
|I||Normal radiograph; changes on magnetic resonance imaging|
|II||Sclerosis (wedged, mottled), osteopenia|
|III||Crescent sign indicating a subchondral fracture|
|IV||Flattening and collapse|
|V||Degenerative changes extend to glenoid|
Presentation and History
Glenohumeral arthritis has no classic or pathognomonic history but typically presents as a progressive loss of shoulder rotary mobility coupled with pain. Some persons report noise, grinding, crepitus, and a feeling of a point of “clunking with release.” Another suggestive history includes instability without a history of dislocation, with the incongruent articular surfaces having a mechanical catch. In addition, patients may report morning stiffness that improves throughout the day, along with pain during sleeping.
Primary glenohumeral arthritis in the young population (i.e., <60 years) is exceedingly rare and warrants an exhaustive history to identify the underlying primary diagnosis. History should include prior episodes of trauma, current and previous medications, family history, prior surgical procedures, and recreational and social factors. In addition, the patient’s desired activity level should be taken into consideration. We have found that the desired activity level is a critical aspect of the evaluation of these patients, because outcomes from certain procedures can be perceived as “catastrophic” if patients are not counseled about the longevity and limitations of implants and their ability to meet patients’ expectations.
Examination of the shoulder starts with evaluation of the cervical spine, with notation of deficits in range of motion and reproduction of pain with provocative maneuvers, such as the Spurling test. Referred pain from the cervical spine is common; patients with cervical spine pathology are often more comfortable with their hand draped over their head. Although referred pain and findings of cervical spine pathology do not exclude shoulder pain, patients suspected of having cervical pathology should be referred for proper care and the possibility excluded from the differential diagnosis.
The shoulder examination begins with visual inspection for symmetry, muscle atrophy, and previous surgical incisions. It is critical that the entire shoulder be visualized, including the entirety of the scapula, to ensure that the patient is examined for subtle but significant pathology such as scapular winging. Next, palpation of the posterior capsule, acromioclavicular joint, greater tuberosity, and biceps groove helps to delineate and isolate glenohumeral pathology. Further evaluation includes both passive and active mobility with careful delineation of the painful arc of motion, while looking for the typical finding of an equal limitation of active and passive range of motion in all planes. Motion is evaluated in forward flexion, abduction (controlling for scapulothoracic motion), external rotation at zero and at 90 degrees of abduction, and internal rotation with regard to the vertebral level reached by the hand behind the back. Findings should be compared with the contralateral (presumably unaffected) side. Pain at the mid range of motion and decreased rotational movements should alert the examiner to the presence of an arthritic process, whereas pain at the terminal range of motion is more typically indicative of impingement, osteophytes, and capsular contraction.
Strength testing and impingement signs aid in the evaluation of rotator cuff pathology. Each tendon of the rotator cuff must be evaluated independently. The supraspinatus is evaluated using the Jobe empty can test, whereas infraspinatus testing is accomplished via external rotation strength with the arm adducted at the side. Lag signs must be elicited if they are present, and the Hornblower sign (external rotation with the shoulder abducted 90 degrees) tests the integrity of the teres minor. In this patient population, we have found that it is critical to evaluate the integrity of the subscapularis, especially in the setting of capsulorrhaphy arthropathy after an open anterior instability procedure, where we have noticed a high rate of this finding. Subscapularis tear is typically characterized by asymmetric hyperexternal rotation with the arm in adduction with a positive belly press (a test for the upper subscapularis) and lift-off (a test for the lower subscapularis). The implications of these findings are critical because they may play a significant role in surgical decision making; for example, consideration of procedures such as a concomitant pectoralis major transfer in the setting of subscapularis insufficiency with arthroplasty may be necessary.
Evaluation of Prior Operative Notes and Arthroscopic Images
A complete evaluation of the dates of individual surgeries, prior imaging, and arthroscopic images should also be undertaken for each patient. Such an evaluation can help differentiate between existing pathology and a new diagnosis. Particular attention to results after prior procedures is imperative to understand if the same pathology is present or if further problems have developed since the prior surgery. Arthroscopic images help provide a firsthand view of the cartilage surfaces and the previously treated pathology. Previous MRI scans and radiographs can help put together a picture to further understand previous and new pathology. A review of the operative report of previous surgeries can help the surgeon understand the exact surgical procedures and difficulties affecting the surgical outcomes. Patients often understand that they have a problem in their shoulder but do not effectively communicate exact pathologies that may help guide future considerations.
Careful examination of the detailed operative note should include notation of implants, intraoperative findings, and difficulties with the previous procedures. Prior implants could limit the ability to access the canal if, for example, a biotenodesis screw has been previously placed for a biceps tenodesis. Placement of metal suture anchors into the glenoid and greater tuberosity also could be limiting. These anchors may scratch sensitive metal and polyethylene components that would lead to early failure of further implants without removal of the previous anchors. In addition, certain glenoid anchors have been thought to cause cyst formation, which may make it difficult to place a glenoid component that is stable on the subchondral surface.
Evaluation of all types of arthritis starts with plain film radiographs. X-ray views should include at a minimum a Grashey anteroposterior view, an axillary lateral view at 45 degrees of abduction, and a scapular Y view ( Fig. 56-4 ). These three views are able to document the position of the humeral head in relation to the glenoid, the presence of osteophytes, quality of the bone, the glenohumeral joint space, and glenoid bone loss. They should be evaluated for humeral head posterior subluxation, which may be the earliest sign of arthritis found on radiographs.
Further evaluation includes advanced modalities such as computed tomography (CT) and MRI. CT arthrography is extremely helpful in evaluating joints, especially in the setting of prior hardware placement. Glenoid morphology, bone stock and quality, and the status of the cartilage and rotator cuff can be well evaluated with a CT scan. A glenoid version can be further assessed by the techniques described by Friedman and colleagues ( Fig. 56-5 ). Walch et al. further described the bony anatomy of posterior glenoid wear in the anteroposterior plane, specifically evaluating for biconcavity, which is best evaluated by CT ( Fig. 56-6 ). MRI is especially helpful in evaluating changes to the subchondral bone and associated soft tissue comorbidities. However, the sensitivity and specificity of evaluating chondral lesions is generally regarded as poor; up to 45% of grade IV chondral lesions may be missed.
Treatment goals for the younger athlete with glenohumeral arthritis are resolution of symptoms and restoration of mobility and function. However, patient expectations must be matched to the durability and longevity of the specific treatment chosen. In fact, in 2008, McCarty et al. demonstrated that the most common reason this population pursued treatment was to attempt to return to their previous level of sporting activity. It is well accepted that older patients have reliable outcomes with shoulder replacement, although patients who have early glenohumeral arthritis and their treating surgeons have difficult decisions to make with regard to implant longevity in relation to their age at implantation; it is well known that this population has worse arthroplasty outcomes compared with older populations. The Neer-type total shoulder arthroplasty (TSA) implants have demonstrated an 84% survival rate at 15 years, although revision surgery in the area of the shoulder is extremely complex and few implants have been able to compensate for the glenoid bone loss that is frequently encountered in these revision settings. In young patients, nonoperative treatments should be thoroughly pursued and exhausted before surgical interventions are contemplated. The mainstay of successful outcomes, whether surgical or nonsurgical, is related to patient education that clearly outlines the natural history and functional process that is present, along with the provision of acute and chronic pain management.
The initial treatment of symptomatic glenohumeral osteoarthritis involves recommendation of activity modification, supervised physical therapy, and a trial of oral nonsteroidal antiinflammatory medications. Physical therapy should include a daily regimen of exercise with a focus on strengthening the periscapular, deltoid, and rotator cuff musculature. Stretching the joint with manual manipulation to help improve the range of motion is important to help with the restrictions caused by cartilage damage and resultant capsular contracture. In addition to these modalities, image-guided or blind injection into the glenohumeral joint with a mixture of corticosteroid and lidocaine is an excellent diagnostic and therapeutic tool. However, such an injection may not provide long-term relief for athletic patients engaging in high-demand sports who continue to participate in athletic endeavors, because their symptoms often return with continued activity. Additionally, off-label use of viscosupplementation with hyaluronic acid injections has been shown to provide partial symptom relief. The relief experienced often involves improvement in overall pain and has been effective in allowing patients to sleep through the night with minimal shoulder pain after treatment.
The diagnosis of glenohumeral arthritis is often made incidentally at the time of arthroscopic intervention aimed at other pathology, unless radiographic changes or underlying conditions are present. However, arthroscopy provides a minimally invasive means to evaluate and treat chondral lesions and concomitant pathology in the shoulder, allowing for early rehabilitation and return to activity in symptomatic patients who have not responded to nonoperative measures. Unfortunately, arthroscopy in this setting remains a temporizing measure that does not alter the underlying disease process; however, it can provide pain relief and functional improvement for a considerable amount of time.
Several authors have reported successful results of arthroscopic debridement for glenohumeral arthritis and isolated cartilage lesions in patients younger than 55 years, and several prominent themes have emerged from this work. Results have been reliable and have correlated with the extent of disease. Risk factors for failure are chondral lesions larger than 2 cm 2 , grade IV bipolar lesions, residual joint space less than 2 mm, and the presence of large osteophytes. Weinstein et al. reported that pain relief and maintenance of function was sustained for 76% of patients at 34-month follow-up, and it was sustained in 88% of patients at 2-year follow-up in the series by Cameron et al. Conversion to arthroplasty was required in 22% of patients at an average of 10.1 months after debridement in one series and in 15% of patients at 20 months in another series.
Several concomitant procedures were performed in the aforementioned series, including bicep tenotomy/tenodesis, distal clavicle resection, subacromial decompression, microfracture, loose body/osteophyte removal, and capsular release. We have found that capsular release is extremely useful, and it has been recommended that capsular release be included in arthroscopic treatment if rotation is limited by greater than 15 degrees in any plane of motion. Significant attention must be given to these aforementioned concomitant pain generators in the setting of arthritic disease, and furthermore, failure to address these pathologies has been demonstrated to produce worse postoperative results.
In addition to debridement, it has been recommended that full-thickness cartilage lesions be managed with use of the marrow stimulation technique of microfracture, similar to that described in the knee ( Fig. 56-7 ). In 2009, Millett et al. reported on 31 patients (average age: 43 years) with full-thickness cartilage lesions in the glenohumeral joint who underwent treatment using microfracture. At a mean follow-up of 47 months, the patients demonstrated significant improvement in the visual analog scale score, American Shoulder and Elbow Surgeons score, and painless use of the arm, as well as improvement in patients’ ability to work, perform activities of daily living, and participate in sports activity, with failure in 19% of patients, who required additional procedures. These investigators demonstrated the greatest improvement in patients with smaller, isolated humeral head lesions and the worst results in patients with bipolar lesions. More recently, Frank et al. published their experience with a similar cohort of 15 shoulders in 14 patients (average age: 37 years) who underwent glenohumeral microfracture and demonstrated similar results. At an average follow-up of 28 months, these investigators showed significant improvements in the visual analog scale, simple shoulder test, and American Shoulder and Elbow Surgeons scores, with 92% of patients stating that they would have the procedure again. The procedure was considered a failure in three patients in this series (20%), who required further operative intervention during the study period.
Glenoidplasty and Osteocapsular Arthroplasty
In concordance with the concept of posterior subluxation and nonconcentric posterior glenoid wear, the technique of glenoidplasty was developed in an effort to restore glenoid morphology from a biconcavity to a single concave surface. Restoration of uniform concavity theoretically reduces posterior humeral head subluxation and increases the effective joint surface, thus reducing point loading while increasing stability. This technique is performed by using an arthroscopic burr where the anterior glenoid (or “high side”) is resected until it is flush with the posterior joint. Recommendations for this procedure are aimed at pain relief; however, it has been demonstrated that patients with pain in their mid-arc range of motion do not fare as favorably after surgery as those who have end–range of motion pain, pain at rest, and painless crepitus. Also associated with favorable results are the radiographic findings of a biconcave glenoid with posterior humeral subluxation and the presence of osteophytes and loose bodies. Contraindications to this procedure are glenoids without biconcavity and patients with severe pain in mid-arc motion, because these findings have been associated with severe arthritis, and patients with these findings have not done well after glenoidplasty.
Osteocapsular arthroplasty involves removal of humeral osteophytes with release of capsular contractures. Complete capsular release from the rotator interval to the posterior-inferior capsular recess should be performed to gain maximal mobility. In a cohort with an average age of 50 years who were treated with this procedure, Kelly et al. demonstrated that more than 85% of patients reported improvement in both pain and range of motion at an average 3-year follow-up. Elser et al. have briefly presented their data on 27 young patients who underwent a similar procedure that they have termed comprehensive arthroscopic management, which includes debridement, capsular release, and humeral osteoplasty. One-year follow-up demonstrated a patient satisfaction rate of 8.5/10 with the procedure, with only one patient progressing to arthroplasty in that period.
Arthroscopy for Rheumatoid Arthritis and Osteonecrosis
Arthroscopy has also proven useful in the early treatment of both rheumatoid arthritis and osteonecrosis. The hypertrophic synovium of rheumatoid disease leads to bony destruction, and thus a synovectomy may help slow the disease progression. Synovectomy results in increased motion and decreased pain in 80% of patients; however, this procedure is only useful early in the disease process, and when radiographic signs of disease are present, synovectomy is no longer a viable option. Concomitant procedures should be performed as indicated, including subacromial decompression and rotator cuff repair, and the synovium may reproliferate, requiring a repeat synovectomy.
Patients with osteonecrosis have also benefited from arthroscopy and core decompression early in their disease process, and results have been shown to correlate to the severity of humeral head involvement. This technique was first described in 1993 by Mont et al. and has been shown to be successful in 94% and 88% of patients with stage I and II disease, respectively. These results decline to 70% and 14% success in patients with stage III and IV disease, respectively. Core decompression may delay the time until arthroplasty in persons with stage III disease, although indications are controversial and this procedure is contraindicated in persons with stage IV and V disease.
Joint Resurfacing Techniques
Humeral Head Resurfacing, Biologic Resurfacing, and Osteochondral Grafting
Secondary treatment options include humeral head resurfacing, biologic resurfacing of the glenoid, and osteochondral allograft or autograft of the humeral head or glenoid. These techniques are used along the spectrum depending on the amount and chronicity of injury to the articular surfaces. Also involved is the level of instability present as a result of underlying pathology. Osteochondral allografts have been suggested as a means for reconstructing defects in both the glenoid and humeral head in cases of instability of the shoulder, particularly in the setting of large Hill-Sachs lesions. In the chronic setting, or when larger humeral defects are present, a structural allograft can be used as a biologic means of extending the range of contact with the glenoid surface in external rotation. Algorithms for treating humeral head defects have been well described. For older patients with more than 45% of the humeral head affected, a prosthetic replacement is recommended. For younger patients with greater than a 45% structural defect, an osteochondral allograft should be considered ( Fig. 56-8 ). Evaluation of outcomes has largely relied on case reports demonstrating success at 2-year follow-up. Given the rarity of these cases and limited reports, each patient should be approached individually and treatment should be determined on a case-by-case basis. Allograft reconstruction techniques for the glenoid have been shown to be beneficial at 5- to 7-year follow-up.
Humeral head resurfacing has been reviewed as a further consideration for patients who are young and active. Since its inception in 1979, this treatment has undergone numerous changes, including the addition of hydroxyapatite coating in 1993. Current techniques including the “ream and run,” designed to promote articular congruity, were first studied by Matsen et al. in a canine model, and demonstrated fibrocartilage formation on the glenoid at 6 months. This model was then transitioned into a human population, and results were reported in 2007. They demonstrated equivalent subjective shoulder test values at 3-year follow-up when compared with TSA. An important note regarding this technique is that the glenoid should be normalized in morphology, and if posterior subluxation is present, the physician should not perform a posterior release and the inferior glenohumeral ligament should be left intact. Thomas et al. in 2005 and Bailie et al. in 2008 again looked at results after isolated humeral resurfacing and found that patients uniformly had improved outcome scores at 3-year follow-up. Thomas et al. looked at patients with an average age of 70 years, whereas Bailie et al. looked at a subset of patients younger than 50 years, which suggests that resurfacing is an acceptable treatment option for patients younger than 50 years who would like to maintain an active lifestyle. Levy and Copeland also looked at a subset of patients younger than 50 years and concluded at an average follow-up of 8.2 years that range of motion and Constant scores were equal to the outcomes of TSA. These studies provide validated outcome measures that give the treating surgeon a viable alternative to traditional shoulder replacement in the younger population for whom the longevity of glenoid fixation may be of concern.
Biologic resurfacing that focuses on glenoid “coverage,” combined with a humeral-sided hemiarthroplasty or resurfacing, has also been proposed as a solution to glenoid longevity and ultimate implant survival in the young, active population. Several glenoid resurfacing techniques have been proposed and studied, including use of a fascia lata autograft, an anterior shoulder capsule, an Achilles allograft, a meniscal allograft, and xenograft tissue ( Fig. 56-9 ). However, the results of these techniques have been mixed. Although the techniques engendered much early enthusiasm because of the initial reports by Nowinski demonstrating that 81% of patients had good or excellent results, with 21 of 26 patients returning to predisease activities (including heavy lifting and manual labor), more recent reports have demonstrated conflicting results. A longer term follow-up study by Krishnan et al. showed excellent results in 18 of 36 patients (50%), satisfactory results in 13 of 36 patients (36%), and unsatisfactory results in 5 of 36 patients (14%) at 2- to 15-year follow-up with use of an anterior capsule, fascia lata, or Achilles allograft. These investigators concluded that use of capsular tissue was an independent predictor of failure, and thus they recommend use of an Achilles allograft in this setting. Similarly, Wirth reported on 27 patients who underwent resurfacing with a lateral meniscal allograft and demonstrated significant improvements in pain relief and function at an average of 3 years of follow-up. However, this group demonstrated uniform progression of joint space narrowing, which calls into concern the long-term functional outcome in this cohort. These mostly positive results must be contrasted with the recent results reported by Elhassan et al. regarding the experience of the Harvard Shoulder Service with a similar procedure in 13 patients, with use of an Achilles allograft (11), a fascia lata allograft (1), and an anterior joint capsule allograft (1). Ten of 13 patients required revision TSA at a mean of 14 months, one had good function and pain relief, and two cases were complicated by infection, one of which was salvaged with antibiotics and irrigation and debridement, with the other patient undergoing resection arthroplasty. Thus this surgical option is attractive because it allows younger, more active patients to return to their previous level of activity, including weight lifting and manual labor, without concern for eccentric glenoid wear, although the long-term results may be unpredictable.
Shoulder arthroplasty was first performed by a French surgeon named Pean in 1893, but it was largely pioneered by Neer in the 1950s, and persistent evolution of prosthetic design and surgical technique has continued to improve function in patients undergoing shoulder arthroplasty. Patients with primary osteoarthritis, posttraumatic arthritis, inflammatory arthritis, osteonecrosis, capsulorrhaphy, and dislocation arthropathy can be effectively treated with shoulder arthroplasty. Pain that has failed to respond to other interventions is the most common indication for arthroplasty; however, patients often obtain statistical improvement with regard to range of motion, strength, and function as well. The status of the soft tissues, bony structures, and underlying etiology must be taken into account when counseling patients about expected results and when planning the details of operative intervention.
Primary osteoarthritis of the glenohumeral joint that is unresponsive to nonoperative or joint-preserving techniques is a well-supported indication for shoulder arthroplasty ( Fig. 56-10 ). Godeneche et al. reported excellent results in a series of 268 patients. Objective outcomes improved based on Constant scores with 77% good or excellent results, motion gains were seen with 50-degree improvements in elevation, and subjective scores demonstrated 94% satisfaction rates. In a large multicenter study of more than 600 patients, Edwards et al. showed clinically and statistically significant improvements for all variables in patients undergoing total and hemiarthroplasty of the shoulder, including Constant scores, pain, range of motion, activity, and strength. Khan et al., in a 10-year prospective evaluation of patients who underwent total shoulder arthroplasty, found a mean improvement of 41 points in Constant scores and a 65-degree improvement in forward elevation for patients with primary osteoarthritis. Component survival was found to be 100% for the humeral component and 92% for the glenoid. Raiss et al. evaluated younger patients with mean age of 55 years who underwent TSA for primary osteoarthritis and found a 40-point increase in Constant scores, 95% satisfaction rates, and 100% component survival at a minimum 5-year follow-up. These results are consistent with the results of other series on arthroplasty for primary osteoarthritis and demonstrate predictable and reproducible improvements in all parameters evaluated, providing thorough evidence when counseling patients.