➢ Reverse shoulder arthroplasty for the treatment of complex fractures of the proximal part of the humerus appears to have good results at short and intermediate-term follow-up.
➢ Malunion or nonunion of the tuberosities does not affect the functional result after reverse total shoulder arthroplasty as much as it does after hemiarthroplasty, but it does lead to decreased postoperative external rotation.
➢ Long-term outcomes of reverse shoulder arthroplasty for the treatment of fractures have still not been well established.
Proximal humeral fractures account for approximately 5% of fractures that are seen in the emergency department and can present difficult challenges to the treating physician1. These fractures are most commonly classified according to the system proposed by Neer2. Complex proximal humeral fractures that require operative treatment can be treated with osteosynthesis, hemiarthroplasty, total shoulder arthroplasty, or, more recently, reverse shoulder arthroplasty3. While locked plating may result in improved fixation in osteopenic bone compared with traditional unlocked plating, the outcomes may still be limited by the risk of complications, including stiffness, osteonecrosis, implant failure, and loss of reduction4,5. Hemiarthroplasty traditionally has been performed for the treatment of three or four-part fractures, particularly in patients with osteopenic bone, and can yield good results for low-demand patients6,7. The outcomes of hemiarthroplasty for the treatment of proximal humeral fracture are satisfactory for long-term pain relief but are less predictable in terms of shoulder motion8. Reverse shoulder arthroplasty can be used for the treatment of cuff tear arthropathy and recently has gained popularity for the treatment of severe proximal humeral fractures. The present review of the current literature on the use of reverse shoulder arthroplasty for the treatment of complex proximal humeral fractures focuses on indications, preoperative planning, surgical techniques, and complications.
Injuries involving the proximal part of the humerus prompted approximately 184,000 visits to the emergency department in 2008 in the United States. The overwhelming majority of these injuries occurred in patients older than fifty years of age, with the number peaking in the eighty to eighty-five-year age range. The incidence increased exponentially between the ages of forty and eighty-four years for women and between the ages of sixty and eighty-nine years for men9. The most common mechanism of fracture in elderly patients is a fall from standing.
A complete history and physical examination, including an assessment of general health status, is necessary when evaluating a patient who has a suspected proximal humeral fracture. Range of motion and strength are usually limited because of pain. A detailed neurovascular examination of the affected extremity should be performed because injuries to the axillary and suprascapular nerves and the axillary artery can occur in association with proximal humeral fractures10,11. In the acute setting, axillary and suprascapular nerve function can be evaluated by testing isometric muscle contraction as well as sensibility in the distribution of the axillary nerve.
Radiographs include a minimum of two orthogonal radiographs of the glenohumeral joint12. A true anteroposterior (Grashey) radiograph and an axillary lateral radiograph can be used to assess the congruity of the glenohumeral joint13. When pain precludes abduction of the arm for an axillary lateral radiograph, a true scapulolateral (scapular-Y) and/or a Velpeau axillary radiograph should be made14,15. Computed tomography (CT) with three-dimensional reconstruction can provide excellent visualization of fracture fragments and also can be used to evaluate for associated fractures of the glenoid and scapular body. Magnetic resonance imaging (MRI) can be used to determine the status of the rotator cuff and to aid in surgical decision-making, particularly when choosing between hemiarthroplasty and reverse shoulder arthroplasty.
The most commonly employed classification system for proximal humeral fractures is the Neer classification system2. This system describes fractures in terms of displacement of four segments of the proximal part of the humerus as identified originally by Codman: the greater tuberosity, lesser tuberosity, head, and shaft16. Displaced fractures include two-part, three-part, and four-part fractures; the definition of a part requires an individual fragment to be displaced by >1 cm or rotated 45° from its normal anatomic position. Minimally displaced fractures, therefore, can have fracture lines through multiple fragments but are classified as one-part fractures. Furthermore, the system classifies articular surface head-splitting fractures as well as fracture-dislocations according to parts and anterior or posterior displacement of the head. The AO Müller classification system divides proximal humeral fractures into three groups (A, B, C) with three separate subgroups (A1, A2, A3, etc.), with consideration being given to the number of fracture lines, the location of fractures, impaction, dislocation, and displacement17.
The limitations of the Neer and AO Müller proximal humeral classification systems rest in their low intraobserver reproducibility and interobserver reliability15; nevertheless, they represent the best attempts at organizing a natural continuum of injury into separate categories with prognostic and treatment implications.
The majority of proximal humeral fractures in elderly patients can be successfully treated nonoperatively with an initial period of immobilization in a sling followed by a gradual return to activities. However, nonoperative treatment of three and four-part fractures often results in a poor functional result because of nonunion, malunion, posttraumatic glenohumeral arthritis, or stiffness18,19. Surgical options include closed reduction and percutaneous pinning, open reduction and internal fixation with locked or unlocked plating, locked intramedullary nailing, or shoulder arthroplasty, depending on the fracture pattern and on patient and surgeon preference.
Adequate reduction and fixation of three and four-part fractures is technically very difficult to obtain and maintain with closed reduction and percutaneous pinning. Open reduction and internal fixation is possible but is associated with a high rate of complications, including screw penetration, loss of fixation, and osteonecrosis of the humeral head, leading to poor outcomes and a failure rate of 15% to 20%20,21.
Neer described the use of hemiarthroplasty to treat three and four-part fractures of the proximal part of the humerus6. Although Neer reported good results, subsequent studies failed to produce predictably good functional outcomes8,22,23. Complications of hemiarthroplasty for the treatment of proximal humeral fracture have included poor range of motion, glenoid erosion, component loosening or migration, and persistent pain. The most common complication associated with a poor outcome after hemiarthroplasty for the treatment of a proximal humeral fracture is malunion or nonunion of the greater tuberosity, which occurs in 39% to 50% of cases3,23-25.
Because of a lack of consistently good outcomes for patients managed with the above-mentioned techniques, surgeons began to explore the efficacy of reverse shoulder arthroplasty for these complicated and difficult fractures. Fractures associated with severe tuberosity comminution may be associated with a higher rate of failure when treated with traditional open reduction and internal fixation or hemiarthroplasty, and, therefore, reverse shoulder arthroplasty may be a better option for these fractures. Reverse shoulder arthroplasty may be indicated for acute, comminuted three or four-part proximal humeral fractures in elderly patients or acute fractures in patients with preexisting rotator cuff deficiency. In addition, reverse shoulder arthroplasty is indicated for the treatment of fracture sequelae in patients who have undergone osteosynthesis that has resulted in rotator cuff or tuberosity deficiency or those who have had a failed hemiarthroplasty that has resulted in tuberosity deficiency. Reverse shoulder arthroplasty should be avoided in younger, active patients as the results have been shown to deteriorate as early as six years postoperatively26. The surgeon should be very familiar with the operative technique of reverse shoulder arthroplasty, and consideration should be given to mastering the technique on a cadaver specimen.
A standard deltopectoral approach with the patient in the beach-chair position allows for exposure of the proximal humeral fracture as well as for identification and mobilization of the greater and lesser tuberosities. Although a superolateral approach frequently is used in the setting of reverse shoulder arthroplasty for the treatment of rotator cuff tear arthropathy, this approach can also be used in the fracture setting, but access to the lesser tuberosity fragment may be difficult.
The rotator interval is delineated, and the greater and lesser tuberosities are mobilized. The greater and lesser tuberosities should be mobilized enough so that they can be retracted out of the way for access to the glenoid. Heavy traction sutures are placed through the greater and lesser tuberosities and the rotator cuff, with any comminuted tuberosity fragments being left attached to the rotator cuff. Any remaining soft-tissue attachment of the lesser tuberosity is left undisturbed. A subpectoral soft-tissue biceps tenodesis is performed with heavy nonabsorbable suture, and the intra-articular biceps tendon stump is resected.
The fractured humeral head is extracted and measured. No further humeral resection is typically needed. The humeral canal is sequentially reamed by hand to prevent iatrogenic fracture propagation. A trial humeral component is placed to confirm proper humeral canal sizing and to approximate humeral height, although proper humeral height cannot be confirmed until after the glenosphere is placed. The humeral trial component is then removed, and attention is turned to glenoid preparation.
The greater tuberosity is retracted posteriorly, and the lesser tuberosity is retracted anteriorly, with care being taken not to detach any remaining soft-tissue attachment. The labrum and the biceps tendon stump are circumferentially excised. The glenoid central drill hole is placed slightly inferior and posterior to the center of the glenoid. The glenoid is then reamed to remove articular cartilage, which is usually present in the setting of acute fractures. The glenoid baseplate is placed with a 10° to 15° tilt from inferior to superior to avoid scapular notching27-29. The proper glenosphere is selected, keeping in mind that a larger glenosphere provides optimal range of motion and stability. When the glenosphere is placed and stability is confirmed, attention is then turned to the humeral side.
It is critical to place the humeral component at the proper height to ensure that proper soft-tissue tensioning is achieved in order to prevent component instability and to maximize range of motion. Humeral height can be established with use of one of several methods, including use of preoperative radiographs of the contralateral humerus to measure anatomic humeral height30 and the relationship of the superior aspect of the humeral head and the upper border of the pectoralis major tendon (approximately 5.6 cm)31,32, with the goal being anatomic reconstruction so that the normal position of the greater tuberosity relative to the humeral shaft can be restored.
Trial humeral stem and polyethylene components are placed. Provisional suture fixation of the tuberosities provides an approximation of appropriate height and proper tension. A trial reduction is performed to further assess proper humeral height and tension. After removal of the trial component and thorough canal irrigation, heavy nonabsorbable sutures are placed in the humerus through drill holes for tuberosity repair. A cemented humeral component should be used as the fractured proximal part of the humerus cannot provide enough osseous stability to stabilize a press-fit humeral stem. The humeral stem is cemented in place in minimal posterior version with use of the markings on the trial component as a guide to the appropriate height. Humeral version can be confirmed with use of a guide pin in the humeral stem handle or by marking the handle with a marking pen. Placing the stem between neutral and 20° of retroversion has been shown to provide optimal stability and to minimize scapular notching33,34. The trial humeral cup is again inserted. The proper tension can be assessed on the basis of the tension of the deltoid and conjoined tendon. The tuberosities are repaired horizontally to each other and to the prosthesis itself or around the medial neck of the prosthesis and then are repaired vertically with transosseous sutures.
Preservation of the tuberosities allows as much rotator cuff function as possible and provides bone stock to support the implant. Bone graft from the humeral head is packed between the tuberosities as well as between the tuberosities and the implant. The rotator interval is usually nonrepairable because of the inferior tension on the humeral component. Intraoperative fluoroscopy with C-arm imaging is used to confirm tuberosity and implant placement as well as concentric glenohumeral reduction as well as the avoidance of intraoperative glenoid or humeral fracture (Fig. 1).
Each step in glenoid and humeral preparation is performed with an eye toward avoiding the most common complications of reverse shoulder arthroplasty, including dislocation, intraoperative fracture, hematoma, infection, and scapular notching10,35-37. Risk factors that contribute to instability after reverse shoulder arthroplasty include poor soft-tissue tension, improper humeral version, and deltoid dysfunction38. Repair of the subscapularis tendon and lesser tuberosity is important given the increased rate of dislocation after reverse shoulder arthroplasty in patients with a nonrepairable subscapularis tendon39. Meticulous hemostasis is achieved, and the wound is closed over a drain. The risk of scapular notching can be reduced by inferior placement and inferior angulation of the glenosphere.
Physical therapy involving gentle passive range of motion is started immediately after surgery. Active abduction and internal and external rotation are avoided for the first six weeks postoperatively. The patient is instructed to absolutely avoid using the affected arm to push up from a seated position, particularly for the first six weeks postoperatively. Active range of motion and strengthening are advanced after tuberosity healing is confirmed radiographically.
Results of Reverse Shoulder Arthroplasty for Proximal Humeral Fracture
Reverse shoulder arthroplasty for the treatment of complex fractures of the proximal part of the humerus appears to have acceptable short and intermediate-term results for as long as six years (Table I and Table II). Patients are typically able to achieve the ultimate functional result by twelve months postoperatively35. Furthermore, malunion or nonunion of the tuberosities does not affect the functional result as much as it does in patients who have undergone hemiarthroplasty, but it does lead to decreased postoperative external rotation10,35. Although some studies have demonstrated superior outcomes after reverse shoulder arthroplasty as compared with hemiarthroplasty, most reports have presented Level-IV evidence40-42, and we are not aware of any prospective randomized trials comparing reverse shoulder arthroplasty with hemiarthroplasty.
A review of the literature revealed five case series10,35-37,43, including one Level-III case-control study10, of patients undergoing reverse shoulder arthroplasty for the treatment of three and four-part fractures of the proximal part of the humerus (Table I). The review also revealed three retrospective reviews40-42, one Level-III retrospective case-control study44, and one Level-II prospective cohort11 in which the outcomes of hemiarthroplasty were compared with those of reverse shoulder arthroplasty for the treatment of complex fractures of the proximal part of the humerus (Table II). Two studies37,40 on reverse shoulder arthroplasty involved patients as young as fifty-eight years old; otherwise, all patients were sixty-five years or older. The duration of follow-up ranged from four months to sixteen years. Reported outcomes data included range of motion; pain scores; complication rates; and outcome scores, including the Constant score45, American Shoulder and Elbow Surgeons (ASES) score46, Oxford Shoulder Score (OSS)47, Simple Shoulder Test (SST) score48, Disabilities of the Arm, Shoulder and Hand (DASH) score49, University of Pennsylvania (Penn) Score50, and Single Assessment Numeric Evaluation (SANE) score51 (Table I and Table II). In addition, Mata-Fink et al.52 performed a systematic review comparing the functional outcomes of reverse shoulder arthroplasty with those of hemiarthroplasty for the treatment of proximal humeral fractures in older adults. The authors concluded that reverse shoulder arthroplasty resulted in improved range of motion and functional outcome scores (ASES, Constant, and Oxford scores), with no difference in the rate of complications between groups. Furthermore, the authors recommended that patients should be counseled that the long-term outcomes of reverse shoulder arthroplasty for the treatment of fractures have still not been well established.
On the basis of the current literature, it appears that reverse shoulder arthroplasty for the treatment of complex fractures of the proximal part of the humerus provides acceptable functional outcomes for patients with three and four-part fractures. In general, the short-term recovery and patient satisfaction after reverse shoulder arthroplasty for the treatment of fracture appear to be at least similar to those after hemiarthroplasty. However, it is important to note that these studies included relatively few patients and involved short and intermediate-term follow-up.
Complications of reverse shoulder arthroplasty include scapular notching, neurologic injury, glenoid fracture, component loosening, dislocation, infection, and hematoma (Table I and Table II). The most common complication after reverse shoulder arthroplasty is scapular notching, which occurred in fifteen of sixteen patients in the study by Gallinet et al.40. The clinical importance of scapular notching is unclear, particularly in the early postoperative period; however, pain and active range of motion have been shown to worsen with scapular notching after five years postoperatively53. Tuberosity-related complications are also common and occurred in nineteen of forty-three patients undergoing reverse shoulder arthroplasty for the treatment of fracture in the study by Bufquin et al.35. Although tuberosity-related complications after reverse shoulder arthroplasty for the treatment of humeral fractures do not affect outcomes as much as they do after hemiarthroplasty, one study comparing tuberosity repair with tuberosity excision demonstrated lower Constant scores and decreased range of motion, most pronounced in external rotation, for patients who had tuberosity excision10. Finally, the study with the longest reported follow-up for patients undergoing reverse shoulder arthroplasty for the treatment of proximal humeral fracture demonstrated that Constant scores decreased over time and scapular notching increased over time, although there was no comparison group37.
The results of reverse shoulder arthroplasty for the treatment of proximal humeral fractures are encouraging. Prospective randomized studies are needed to compare the outcomes of reverse shoulder arthroplasty with those of hemiarthroplasty in this scenario. Longer follow-up documenting component longevity and late complications is also necessary.
Source of Funding: No funds were received in support of this study.
Investigation performed at KSF Orthopaedic Center, Houston, Texas, and University of Texas Southwestern Medical Center, Dallas, Texas
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2014 by The Journal of Bone and Joint Surgery, Incorporated